Pasar al contenido principal

ENTORNOS 3D DE ALTA FIDELIDAD PARA REALIDAD VIRTUAL Y COMPUTACION VISUAL: GEOMETRIA, MOVIMIENTO, INTERACCION Y VISUALIZACION PARA SALUD, ARQUITECTURA Y CIUDADES

PID2021-122136OB-C21

•
Nombre agencia financiadora Agencia Estatal de Investigación
Acrónimo agencia financiadora AEI
Programa Programa Estatal para Impulsar la Investigación Científico-Técnica y su Transferencia
Subprograma Subprograma Estatal de Generación de Conocimiento
Convocatoria Proyectos de I+D+i (Generación de Conocimiento y Retos Investigación)
Año convocatoria 2021
Plan Estatal 2021-2023 Plan Estatal de Investigación Científica y Técnica y de Innovación
Centro beneficiario UNIVERSITAT POLITECNICA DE CATALUNYA

Resultados relacionados

Resultados totales (Incluyendo duplicados): 108
Encontrada(s) 2 página(s)

Repositorio Digital de la UPF
Publicación Artículo científico (JournalArticle)

A highly-configurable session designer for VR nursing training

  • Andújar, Carlos
  • Chica, Antoni
  • Fairén, Marta
  • García, Óscar
  • Nieto, Jordi
  • Tortosa, Sergi
  • Insa-Calderón, Esther
Virtual Reality (VR) has proven to be a valuable tool for medical and nursing education. VR simulators are available at any time and from anywhere, and can be used with or without faculty supervision, which results in a significant optimization of time, space, and resources. In this paper we present a highly-configurable session designer for VR-based nursing education following the Standards for QUality Improvement Reporting Excellence: SQUIRE 2.0 and SQUIRE-EDU. Unlike existing platforms, we focus on letting educators quickly customize the training sessions in multiple aspects. This is achieved through the use of a visual editor, where educators can choose the environment and the elements in the session, and a node-based visual system for programming the behavior of the sessions. Thanks to end-user customization, educators can develop different variants of a session, adapt the sessions to new nursing protocols, incorporate new equipment or instruments, or accommodate custom environments. We have implemented the designer and tested it on various nursing training scenarios. The designer can be integrated into VR simulators to help educators save time in delivering highly customizable training., This research was partially supported by MICIU / AEI / 10.13039/501100011033 and ERDF / EU under Grant PID2021-122136OB-C21.
Repositori Obert UdL
Publicación Artículo científico (JournalArticle)

Padelvic: Multicamera videos and motion capture data in padel matches, PADELVIC: Videos multicámara y datos de captura de movimiento de un partido de pádel amateur

  • Javadiha, Mohammadreza
  • Andujar, Carlos
  • Calvanese, Michele
  • Lacasa Claver, Enrique
  • Moyés, Jordi
  • Pontón, José Luis
  • Susin, Antonio
  • Wang, Jiabo
Los recientes avances en visión por computador y las técnicas de aprendizaje profundo han abierto nuevas posibilidades en cuanto a la etiquetación automática de videos deportivos. Sin embargo, un requisito esencial para las técnicas de aprendizaje supervisado es la disponibilidad de extensos conjuntos de entrenamiento con anotaciones precisas. En este artículo presentamos PadelVic, un conjunto de datos anotado de un partido amateur de pádel que incluye múltiples secuencias de video desde diferentes ángulos, datos precisos de captura de movimiento de uno de los jugadores, así como videos sintéticos diseñados específicamente para ser utilizados como conjuntos de entrenamiento para redes neuronales convolucionales que estiman datos de posición a partir de videos. Como demostración de una de las aplicaciones del conjunto de datos, presentamos un sistema para predecir con precisión el centro de masa de los jugadores proyectado sobre el plano de la pista, a partir de un solo video del partido., Recent advances in computer vision and deep learning techniques have opened new possibilities regarding the automatic labeling of sport videos. However, an essential requirement for supervised techniques is the availability of accurately labeled training datasets. In this paper we present PadelVic, an annotated dataset of an amateur padel match which consists of multi-view video streams, estimated positional data for all four players within the court (and for one of the players, accurate motion capture data of his body pose), as well as synthetic videos specifically designed to serve as training sets for neural networks estimating positional data from videos. For the recorded data, player positions were estimated by applying a state-of-the-art pose estimation technique to one of the videos, which yields a relatively small positional error (M=16 cm, SD=13 cm). For one of the players, we used a motion capture system providing the orientation of the body parts with an accuracy of 1.5º RMS. The highest accuracy though comes from our synthetic dataset, which provides ground-truth positional and pose data of virtual players animated with the motion capture data. As an example application of the synthetic dataset, we present a system for a more accurate prediction of the center-of-mass of the players projected onto the court plane, from a single-view video of the match. We also discuss how to exploit per-frame positional data of the players for tasks such as synergy analysis, collective tactical analysis, and player profile generation., This work has received funding from project PID2021-122136OB-C21 funded by MCIN/AEI and FEDER “A way to make Europe”. Mohammadreza Javadiha and Jose Luis Ponton were also funded by the Spanish Ministry of Science, Innovation and Universities, grants PRE2018-086835 and FPU21/01927
Repositori Obert UdL
Publicación Artículo científico (JournalArticle)

A Query Language for Exploratory Analysis of Video-Based Tracking Data in Padel Matches

  • Javadiha, Mohammadreza
  • Andujar, Carlos
  • Lacasa Claver, Enrique
Recent advances in sensor technologies, in particular video-based human detection, object tracking and pose estimation, have opened new possibilities for the automatic or semi-automatic per-frame annotation of sport videos. In the case of racket sports such as tennis and padel, state-of-the-art deep learning methods allow the robust detection and tracking of the players from a single video, which can be combined with ball tracking and shot recognition techniques to obtain a precise description of the play state at every frame. These data, which might include the court-space position of the players, their speeds, accelerations, shots and ball trajectories, can be exported in tabular format for further analysis. Unfortunately, the limitations of traditional table-based methods for analyzing such sport data are twofold. On the one hand, these methods cannot represent complex spatio-temporal queries in a compact, readable way, usable by sport analysts. On the other hand, traditional data visualization tools often fail to convey all the information available in the video (such as the precise body motion before, during and after the execution of a shot) and resulting plots only show a small portion of the available data. In this paper we address these two limitations by focusing on the analysis of video-based tracking data of padel matches. In particular, we propose a domain-specific query language to facilitate coaches and sport analysts to write queries in a very compact form. Additionally, we enrich the data visualization plots by linking each data item to a specific segment of the video so that analysts have full access to all the details related to the query. We demonstrate the flexibility of our system by collecting and converting into readable queries multiple tips and hypotheses on padel strategies extracted from the literature., This research was funded by the Spanish Ministry of Science and Innovation and FEDER funds, grant number PID2021-122136OB-C21, MCIN/AEI/10.13039/501100011033/FEDER, UE.
Repositori Obert UdL
Publicación Contribución de congreso (ConferenceOutput)

Enriching Medieval 4D Models with Interactive Narratives

  • Andujar, Carlos
  • Cayuela Vellido, Begoña
  • Lorés, Immaculada
  • Bosch Geli, Carles
  • Guardia Pons, Milagros
  • Muñoz-Pandiella, Imanol
Digital technologies are increasingly used in cultural heritage preservation and dissemination, but some monuments pose significant challenges due to substantial historical architectural and pictorial changes. This paper introduces a method to enrich multi-layered, multi-phase 3D models of cultural heritage with art history knowledge. This is achieved through imagebased annotations, interactive narratives, points of interest, text, and audio, creating guided tours of the monument’s historical phases. The approach is user-friendly for art historians, requiring no programming skills for defining such elements. A pilot user evaluation demonstrated its effectiveness in enhancing visitor engagement and understanding, and allowed improving the final web-based application and narratives., This work has been partially supported by projects PID2021-122136OB-C21 and PID2021-122136OB-C22 funded by MCIN/AEI/ 10.13039/501100011033 and ERDF "A way of making Europe", by the EU Horizon 2020, JPICH Conservation, Protection and Use initiative (JPICH-0127), the Spanish Agencia Estatal de Investigación (grant PCI2020-111979) and by the Department of Research and Universities of the Government of Catalonia (2021 SGR 01035).
Repositori Obert UdL
Publicación Artículo científico (JournalArticle)

Revealing the Layers of History: Enriching Medieval Cultural Heritage 4D Models with Interactive Narratives

  • Andujar, Carlos
  • Cayuela Vellido, Begoña
  • Lorés, Immaculada
  • Bosch Geli, Carles
  • Guardia Pons, Milagros
  • Muñoz-Pandiella, Imanol
Cultural heritage (CH) preservation is increasingly leveraging digital technologies to support art history research and provide educational experiences to the public. In the case of some Medieval monuments, the myriad of architectural and pictorial transformations suffered due to the passage of time, and the geographical dispersion of its remaining decoration substantially hinders their understanding. In this article, we present a novel approach for enriching multi-layered, multi-phase 3D models of CH monuments with art history knowledge. Such knowledge is presented in the form of image-based annotations, interactive narratives, points of interest, text, and audio, organized into a collection of guided tours enabling visitors to traverse through different historical phases of the monument. A key benefit of our approach is that art historians require no programming skills to easily define all the elements of the interactive narratives. Our approach integrates various multimedia elements seamlessly into 3D models, offering visitors an engaging and informative journey through space and time. We conducted a pilot user evaluation to demonstrate the effectiveness of our approach, highlighting its value in enhancing visitor engagement and understanding of CH. The study also provided valuable feedback on the user experience, which allowed us to enhance the final version of the web-based application and the associated narratives., This work has been partially supported by projects PID2021-122136OB-C21 and PID2021-122136OB-C22 funded by MCIN/ AEI/10.13039/501100011033 and ERDF “A way of making Europe,” by the EU Horizon 2020, JPICH Conservation, Protection and Use initiative (JPICH-0127), the Spanish Agencia Estatal de Investigación (grant PCI2020-111979), and the Department of Research and Universities of the Government of Catalonia (2021 SGR 01035).
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

Enriching medieval 4D models with interactive narratives

  • Andújar Gran, Carlos Antonio|||0000-0002-8480-4713
  • Cayuela Vellido, Begoña
  • Lorès Otzet, Immaculada
  • Bosch Geli, Carles
  • Guardia Pons, Milagros
  • Muñoz Pandiella, Imanol|||0000-0002-4781-7574
Digital technologies are increasingly used in cultural heritage preservation and dissemination, but some monuments pose significant challenges due to substantial historical architectural and pictorial changes. This paper introduces a method to enrich multi-layered, multi-phase 3D models of cultural heritage with art history knowledge. This is achieved through imagebased annotations, interactive narratives, points of interest, text, and audio, creating guided tours of the monument's historical phases. The approach is user-friendly for art historians, requiring no programming skills for defining such elements. A pilot user evaluation demonstrated its effectiveness in enhancing visitor engagement and understanding, and allowed improving the final web-based application and narratives., This work has been partially supported by projects PID2021-122136OB-C21 and PID2021-122136OB-C22 funded by MCIN/AEI/ 10.13039/501100011033 and ERDF "A way of making Europe", by the EU Horizon 2020, JPICH Conservation, Protection and Use initiative (JPICH-0127), the Spanish Agencia Estatal de Investigación (grant PCI2020-111979) and by the Department of Research and Universities of the Government of Catalonia (2021 SGR 01035)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

Bidirectional exploration of 3D cultural heritage models and photo collections

  • Borràs Civil, Bernat
  • Muñoz Pandiella, Imanol|||0000-0002-4781-7574
  • Andújar Gran, Carlos Antonio|||0000-0002-8480-4713
Advances in 3D digitization have greatly improved the availability of high-quality cultural heritage datasets, often consisting of 3D models and collections of high-resolution photos. However, navigating these 3D datasets poses challenges for art history researchers, especially when they need to repeatedly locate points of interest, get close-up views of them, and analyze their spatial relationships. Existing applications for 3D model sharing often fail to provide a convenient interface to simultaneously explore the 3D model and associated photo collections. This paper proposes a web-based approach that combines a 3D view of the model with a 2D view of a group of photos. Designers can quickly arrange groups of images into 2D projections, and cultural heritage users can explore the 3D model and, simultaneously, the active group of images in the 2D view. Both views are interlinked, so that navigating the 3D model modifies the active photo group, and clicking an image in the 2D view moves the 3D camera to the 3D position the image was taken from. We tested our approach with a 3D dataset on medieval mural paintings. Our tests show that our method helps users quickly locate, inspect and relate different parts of the model, reducing navigation time and frustration. The source code is available at https://github.com/BernatBC/visualitzador-de-fotografies-en-3D., This work has been partially supported by the project PID2021-122136OBC21 funded by MCIN/AEI/ 10.13039/501100011033 and ERDF "A way of making Europe", and by the Department of Research and Universities of the Government of Catalonia (2021 SGR 01035)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Individualized pathfinding in rugged open terrains using semantic navigation meshes

  • Creus López, Carles
  • Pelechano Gómez, Núria|||0000-0002-1437-245X
  • Argudo Medrano, Óscar|||0000-0003-3943-1839
Traditional navigation meshes split the terrain into convex walkable regions, which work well for nearly flat environments with clearly defined obstacles. However, in complex outdoor terrains with variable slopes and heights, these approaches fail to capture areas with consistent characteristics. Moreover, human preferences vary per individual: some may favour smooth paths with gentle slopes, while others prefer shorter routes even if steeper. We present a semantic navigation mesh that generates cells with coherent, low-variance values for chosen semantics, such as slope or vegetation density. Cells are constructed via a region-growing partitioning algorithm, optionally refined for compactness, and can incorporate multiple semantic dimensions simultaneously. This approach preserves terrain complexity without relying on a binary walkable/non-walkable classification. From this partition, we propose different alternatives for generating a navigation graph and perform semantically-aware pathfinding for agent-specific cost functions, which prevents the use of precomputed hierarchical paths. Our evaluation demonstrates that semantic meshes accelerate pathfinding on complex open terrains while producing trajectories close to those obtained by a brute-force search over the entire domain. This provides a flexible, generic framework that balances graph size, subdomain homogeneity, search efficiency, and path quality., This work is part of the project SENDA TED2021-129761B-I00, funded by MCIN/AEI/10.13039/501100011033 and the European Union NextGenerationEU/PRTR, and grant PID2021-122136OB-C21 funded by MICIU/AEI/10.13039/501100011033 and ERDF/EU. The authors also acknowledge support from the Department of Research and Universities of the Government of Catalonia (2021 SGR 01035)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Evaluating LLMs’ abilities to create charts, a systematic approach

  • Ribalta Albado, Maria
  • Vázquez Alcocer, Pere Pau|||0000-0003-4638-4065
The use of generative models, especially those based on pretrained transformers, has become a common practice in code development. Tools such as GitHub Copilot, Cursor, and the direct use of conversational chatbots have proven useful to accelerate the development of applications. Unfortunately, generative models are unable to determine what is correct or wrong, and their outputs may contain errors. Their stochastic nature does not guarantee a single solution for the same problem, either. Furthermore, the output depends largely on the prompt issued by the user. To assess the capabilities of LLMs, some benchmarks have been proposed. Unfortunately, they often rely on ground truth data that may not be available. As a result, the extent to which modern LLMs can create charts needs further investigation. This work contributes to the understanding of the generative models’ ability to create charts in three ways: (a) Creating a dataset of prompts, data sources, and chart types to analyze, (b) Designing a set of systematic experiments that cover a wide range of commonly used charts, and variations of the visual variables, and (c) by empirically analyzing the performance of a large set of LLMs of different sizes, including Claude, CodeLlama, Gemini, Gemma, GPT4o, Llama 3.1, and Mixtral. Our results indicate that even the most advanced LLMs have room for improvement., This project has been supported by PID2021-122136OB-C21 from the Ministerio de Ciencia e Innovación, Spain, by 839 FEDER (EU) funds, and 2021 SGR 01035 by Generalitat de Catalunya, Spain., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

Effect of color palettes in heatmaps perception: a study

  • Molina López, Elena|||0000-0003-4561-6215
  • Middel Soria, Carolina
  • Vázquez Alcocer, Pere Pau|||0000-0003-4638-4065
Heatmaps are a widely used technique in visualization. Unfortunately, they have not been investigated in depth and little is known about the best parameterizations so that they are properly interpreted. The effect of different palettes on our ability to read values is still unknown. To address this issue, we conducted a user study, in which we analyzed the effect of two commonly used color palettes, Blues and Viridis, on value estimation and value search. As a result, we provide some suggestions for what to expect from the heatmap configurations analyzed., This paper has been supported by PID2021-122136OB-C21 from the Ministerio de Ciencia e Innovación, by 839 FEDER (EU) funds. Elena Molina has been supported by FI-SDUR doctoral grant from Generalitat de Catalunya, and FPU grant from the Ministerio de Ciencia e Innovación., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Step2Motion: Locomotion reconstruction from pressure sensing insoles

  • Pontón Martínez, José Luis|||0000-0001-6576-4528
  • Alvarado Pinero, Eduardo
  • Foo, LinGeng|||0000-0002-6082-6002
  • Pelechano Gómez, Núria|||0000-0002-1437-245X
  • Andújar Gran, Carlos Antonio|||0000-0002-8480-4713
  • Habermann, Marc|||0000-0003-3899-7515
Human motion is fundamentally driven by continuous physical interaction with the environment. Whether walking, running, or simply standing, the forces exchanged between our feet and the ground provide crucial insights for understanding and reconstructing human movement. Recent advances in wearable insole devices offer a compelling solution for capturing these forces in diverse, real-world scenarios. Sensor insoles pose no constraint on the users' motion (unlike mocap suits) and are unaffected by line-of-sight limitations (in contrast to optical systems). These qualities make sensor insoles an ideal choice for robust, unconstrained motion capture, particularly in outdoor environments. Surprisingly, leveraging these devices with recent motion reconstruction methods remains largely unexplored. Aiming to fill this gap, we present Step2Motion, the first approach to reconstruct human locomotion from multi-modal insole sensors. Our method utilizes pressure and inertial data—accelerations and angular rates—captured by the insoles to reconstruct human motion. We evaluate the effectiveness of our approach across a range of experiments to show its versatility for diverse locomotion styles, from simple ones like walking or jogging up to moving sideways, on tiptoes, slightly crouching, or dancing. The complete source code, trained model, data, and supplementary material used in this paper can be found at: https://vcai.mpi-inf.mpg.de/projects/Step2Motion/, This work has received funding from MCIN/AEI/10.13039/501100011033/FEDER, UE (Spain) in the framework of the project PID2021-122136OB-C21, and with the support of the Department of Research and Universities of the Government of Catalonia (2021 SGR 01035). Jose Luis Ponton was also funded by the Spanish Ministry of Universities (FPU21/01927 andEST24/00555)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Rendering piecewise approximations of SDFs through analytic intersections

  • Pujol Puig, Eduard|||0000-0001-6406-7499
  • Chica Calaf, Antonio|||0000-0003-0270-2332
Signed distance fields (SDFs) have emerged as an alternative shape representation for real-time collision detection and lighting effects. Computing these for complex models can be expensive, so one popular approach is to prepare an approximation via sampling and interpolation. Then, these may be rendered using sphere marching, which gets close to the surface quickly, but needs several iterations to converge to it. In this paper, we propose an alternative that computes the intersection of a given ray and the surface analytically at a narrow band. This may be combined with other enhancements like having variable error for the approximation depending on the distance to the surface and skipping regions that do not contain the surface to accelerate the outer band ray traversal while reducing the required memory. To achieve smoother representations with minimal computational cost, we propose a method for computing surface intersections and normals from separate interpolants. We evaluate all these to find the optimal combination improving the rendering performance and memory consumption of these SDF approximations., This work has been partially funded by Ministeri de Ciència i Innovació(MICIN), Agencia Estatal de Investigación (AEI) and the Fons Europeu de Desenvolupament Regional (FEDER) (project PID2021-122136OB-C21 funded by MCIN/AEI/10.13039/501100011033/ FEDER, UE); and by the Department of Research and Universities of the Government of Catalonia (2021 SGR 01035). The first author gratefully acknowledges the Universitat Politècnica de Catalunya and Banco Santander for the financial support of his predoctoral grant FPI UPC grant., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

Are LLMs ready for visualization?

  • Vázquez Alcocer, Pere Pau|||0000-0003-4638-4065
© 2024 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., Generative models have received a lot of attention in many areas of academia and the industry. Their capabilities span many areas, from the invention of images given a prompt to the generation of concrete code to solve a certain programming issue. These two paradigmatic cases fall within two distinct categories of requirements, ranging from "creativity" to "precision", as characterized by Bing Chat, which employs ChatGPT-4 as its backbone. Visualization practitioners and researchers have wondered to what end one of such systems could accomplish our work in a more efficient way. Several works in the literature have utilized them for the creation of visualizations. And some tools such as Lida, incorporate them as part of their pipeline. Nevertheless, to the authors’ knowledge, no systematic approach for testing their capabilities has been published, which includes both extensive and in-depth evaluation. Our goal is to fill that gap with a systematic approach that analyzes three elements: whether Large Language Models are capable of correctly generating a large variety of charts, what libraries they can deal with effectively, and how far we can go to configure individual charts. To achieve this objective, we initially selected a diverse set of charts, which are commonly utilized in data visualization. We then developed a set of generic prompts that could be used to generate them, and analyzed the performance of different LLMs and libraries. The results include both the set of prompts and the data sources, as well as an analysis of the performance with different configurations., Supported by Ministerio de Ciencia e Innovación/AEI (PID2021-122136OB-C21 by 10.13039/501100011033/FEDER, UE)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

Exploring the role of expected collision feedback in crowded virtual environments

  • Yun, Haoran
  • Pontón Martínez, José Luis|||0000-0001-6576-4528
  • Beacco Porres, Alejandro|||0000-0001-8192-1431
  • Andújar Gran, Carlos Antonio|||0000-0002-8480-4713
  • Pelechano Gómez, Núria|||0000-0002-1437-245X
An increasing number of virtual reality applications require environments that emulate real-world conditions. These environments often involve dynamic virtual humans showing realistic behaviors. Understanding user perception and navigation among these virtual agents is key for designing realistic and effective environments featuring groups of virtual humans. While collision risk significantly influences human locomotion in the real world, this risk is largely absent in virtual settings. This paper studies the impact of the expected collision feedback on user perception and interaction with virtual crowds. We examine the effectiveness of commonly used collision feedback techniques (auditory cues and tactile vibrations) as well as inducing participants to expect that a physical bump with a real person might occur, as if some virtual humans actually correspond to real persons embodied into them and sharing the same physical space. Our results indicate that the expected collision feedback significantly influences both participant behavior—encompassing global navigation and local movements—and subjective perceptions of presence and copresence. Specifically, the introduction of a perceived risk of actual collision was found to significantly impact global navigation strategies and increase the sense of presence. Auditory cues had a similar effect on global navigation and additionally enhanced the sense of copresence. In contrast, vibrotactile feedback was primarily effective in influencing local movements., This work has received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 860768 (CLIPE project), and from MCIN/AEI/10.13039/501100011033/FEDER, UE (PID2021- 122136OB-C21), ”A way to make Europe”. Jose Luis Ponton was also funded by the Spanish Ministry of Universities (FPU21/01927)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Environment-aware motion matching

  • Pontón Martínez, José Luis|||0000-0001-6576-4528
  • Andrews, Sheldon|||0000-0001-9776-117X
  • Andújar Gran, Carlos Antonio|||0000-0002-8480-4713
  • Pelechano Gómez, Núria|||0000-0002-1437-245X
Interactive applications demand believable characters that respond naturally to dynamic environments. Traditional character animation techniques often struggle to handle arbitrary situations, leading to a growing trend of dynamically selecting motion-captured animations based on predefined features. While Motion Matching has proven effective for locomotion by aligning to target trajectories, animating environment interactions and crowd behaviors remains challenging due to the need to consider surrounding elements. Existing approaches often involve manual setup or lack the naturalism of motion capture. Furthermore, in crowd animation, body animation is frequently treated as a separate process from trajectory planning, leading to inconsistencies between body pose and root motion. To address these limitations, we present Environment-aware Motion Matching, a novel real-time system for full-body character animation that dynamically adapts to obstacles and other agents, emphasizing the bidirectional relationship between pose and trajectory. In a preprocessing step, we extract shape, pose, and trajectory features from a motion capture database. At runtime, we perform an efficient search that matches user input and current pose while penalizing collisions with a dynamic environment. Our method allows characters to naturally adjust their pose and trajectory to navigate crowded scenes., This work has received funding from MCIN/AEI/10.13039/501100011033/FEDER, UE (Spain) in the framework of the project PID2021-122136OB-C21, and with the support of the Department of Research and Universities of the Government of Catalonia (2021 SGR 01035). Jose Luis Ponton was funded by the Spanish Ministry of Universities (FPU21/01927). We thank the 2025 Bellairs Workshop on Computer Animation, where the initial ideas for this research took shape., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

SynReconEval: Synthetic dataset creator for photogrammetric reconstruction evaluation

  • Garrido Melis, Pol
  • Pujol Puig, Eduard|||0000-0001-6406-7499
  • Chica Calaf, Antonio|||0000-0003-0270-2332
SynReconEval is an open-source framework for generating synthetic datasets to benchmark photogrammetric and neural 3D reconstruction methods. It leverages Blender to render controlled RGB images, depth maps, and normal maps with full camera parameters, which are then processed by reconstruction pipelines. The reconstructed models are evaluated against ground-truth geometry using a wide range of metrics, including precision, recall, and spatial error distributions. The system’s modular design allows for easy integration of new renderers, algorithms, and evaluation criteria, enabling large-scale, reproducible, and customizable assessment of reconstruction performance., This work has been partially funded by Ministeri de Ciència i Innovació (MICIN), Agencia Estatal de Investigación (AEI) and the Fons Europeu de Desenvolupament Regional (FEDER) (project PID2021-122136OB-C21 funded by MCIN/AEI/10.13039/501100011033/FEDER, UE); and by the Department of Research and Universities of the Government of Catalonia (2021 SGR 01035). The first author gratefully acknowledges the Universitat Politècnica de Catalunya and Banco Santander for the financial support of his predoctoral grant FPI-UPC grant. Open Access funding was enabled and organized by Universitat Politècnica de Catalunya., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

How to represent landmark trees in digital 3D maps? An automated workflow and user study

  • Grammatikaki, Angeliki
  • Eschner, Johannes
  • Ledermann, Florian
  • Argudo Medrano, Óscar|||0000-0003-3943-1839
  • Waldner, Manuela
Digital 3D maps created from digital elevation models (DEMs) cannot properly capture trees due to the 2.5D nature of the DEMs. Leveraging publicly available DEMs and orthophotos as the only input data sources, we present a fully automatic pipeline that models landmark trees. We conducted two crowdsourced user studies to evaluate visual appeal and scene recognizability using two different levels of detail of tree representations generated using our pipeline. Users found highly detailed trees much more appealing, and also easier and more trustworthy to recognize, with open-ended responses revealing key factors like realism, coherence, and tree shape influencing their preferences. However, the ability to recognize a location seems to depend more on the surrounding environment than the representation of the landmark tree in focus. We discuss the implications of these results for digital 3D outdoor maps., The authors acknowledge TU Wien Bibliothek for financial support through its Open Access Funding Programme. O. Argudo also acknowledges Grant PID2021-122136OB-C21, funded by MICIU/AEI/10.13039/501100011033 and ERDF/EU, as well as the support of the Department of Research and Universities of the Government of Catalonia [2021 SGR 01035]., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Features that influence bike sharing demand

  • Cortez Ordóñez, Alexandra Piedad|||0000-0002-0016-5388
  • Vázquez Alcocer, Pere Pau|||0000-0003-4638-4065
  • Sánchez Espigares, Josep Anton|||0000-0001-8195-1913
During the last few years, Bike Sharing Systems (BSS) have become a popular means of transportation in several cities across the world, owing to their low costs and associated advantages. Citizens have adopted these systems as they help improve their health and contribute to creating more sustainable cities. However, customer satisfaction and the willingness to use the systems are directly affected by the ease of access to the docking stations and finding available bikes or slots. Therefore, system operators and managers' major responsibilities focus on urban and transport planning by improving the rebalancing operations of their BSS. Many approaches can be considered to overcome the unbalanced station problem, but predicting the number of arrivals and departures at the docking stations has been proven to be one of the most efficient. In this paper, we study the features that influence the prediction of bikes' arrivals and departures in Barcelona BSS, using a Random Forest model and a one-year data period. We considered features related to the weather, the stations' characteristics, and the facilities available within a 200-meter diameter of each station, called spatial features. The results indicate that features related to specific months, as well as temperature, pressure, altitude, and holidays, have a strong influence on the model, while spatial features have a small impact on the prediction results., This project has been supported by PID2021-122136OB-C21 from the Ministerio de Ciencia e Innovación, by 839 FEDER (EU) funds and 2021SGR00613-(ADBD) by Generalitat de Cataluña., Peer Reviewed, Objectius de Desenvolupament Sostenible::9 - Indústria, Innovació i Infraestructura, Objectius de Desenvolupament Sostenible::11 - Ciutats i Comunitats Sostenibles
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

A highly-configurable session designer for VR nursing training

  • Andújar Gran, Carlos Antonio|||0000-0002-8480-4713
  • Chica Calaf, Antonio|||0000-0003-0270-2332
  • Fairén González, Marta|||0000-0001-7293-584X
  • Garcia Toledo, Òscar
  • Nieto Maldonado, Jordi
  • Tortosa Ortiz-Villajos, Sergi
  • Insa Calderón, Esther
Virtual Reality (VR) has proven to be a valuable tool for medical and nursing education. VR simulators are available at any time and from anywhere, and can be used with or without faculty supervision, which results in a significant optimization of time, space, and resources. In this paper we present a highly-configurable session designer for VR-based nursing education following the Standards for QUality Improvement Reporting Excellence: SQUIRE 2.0 and SQUIRE-EDU. Unlike existing platforms, we focus on letting educators quickly customize the training sessions in multiple aspects. This is achieved through the use of a visual editor, where educators can choose the environment and the elements in the session, and a node-based visual system for programming the behavior of the sessions. Thanks to end-user customization, educators can develop different variants of a session, adapt the sessions to new nursing protocols, incorporate new equipment or instruments, or accommodate custom environments. We have implemented the designer and tested it on various nursing training scenarios. The designer can be integrated into VR simulators to help educators save time in delivering highly customizable training., This research was partially supported by MICIU / AEI / 10.13039/501100011033 and ERDF / EU under Grant PID2021- 122136OB-C21., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Improved multi-label hierarchical patent classification using LLMs

  • Rafieian, Bardia
  • Vázquez Alcocer, Pere Pau|||0000-0003-4638-4065
Classifying multi-label documents has always been a challenging task, especially when the labels follow a hierarchical structure. This complexity increases the difficulty of accurately predicting multiple interrelated labels, often limiting the overall classification performance. To address these challenges and improve the accuracy of hierarchical multi-label classification systems, we introduce a novel pipeline leveraging large language models (LLMs). By incorporating quantization techniques and optimizing weight updates through smaller matrices, we improve computational efficiency and scalability. Our approach demonstrates an improvement in accuracy compared to current state-of-the-art models. In this work, we apply this pipeline to patent documents, focusing on the multi-label hierarchical text classification problem using a transformer-based architecture. The hierarchical structure of the CPC (Cooperative Patent Classification) labels is preserved through a graph-based taxonomy, which enables more effective processing of patent categories. Our model is trained and evaluated on the USPTO-70k dataset, and we achieve substantial improvements across various metrics, including precision, recall, F1-score, and AUC., This project has been supported by PID2021-122136OB-C21 from the Ministerio de Ciencia e Innovación, by 839 FEDER (EU) funds., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Social crowd simulation: improving realism with social rules and gaze behavior

  • Itatani, Reiya|||0000-0002-8430-1657
  • Pelechano Gómez, Núria|||0000-0002-1437-245X
Current crowd simulation models focus mostly on steering towards a goal while avoiding collisions based on the agent’s direction of movement. This leads to robot-like simulations since agents’ appear to always have their attention perfectly aligned with the direction of movement. In the real world, we observe that humans move in a crowd performing collision avoidance driven by attention, gaze, and non-verbal coordination with incoming traffic. In addition, humans exhibit different steering strategies based on whether they walk alone or in a group, whether they can look ahead and plan their best local movement, or react more abruptly because their gaze diverts from their direction of movement. Human gaze can be driven by movement, but also by distractions such as being engaged in conversation with other people or using mobile phones. These human features are overlooked in crowd simulation, often leading to perfectly smooth local movements of individuals. Unfortunately, this lack of social behaviour and variety in animations may be perceived as unrealistic when observing the results on a 2D display, and it may become even more apparent in immersive scenarios where the participant is at eye level with the virtual humans. This paper proposes a novel approach to enhance the realism of a rule-based crowd simulation model by incorporating social rules and gaze-driven attention with consistent animations. The ultimate goal is to make immersive virtual crowds more realistic. Our proposed method enhances existing crowd simulation frameworks by integrating social behavior models that affect both individual and collective dynamics, and drives gaze behavior to better simulate attention. We conducted validation user studies on both a 2D display and in immersive VR, and observed that applying these models to both the steering and animation levels significantly improves the realism of the crowd simulation. The 2D display based user study based on video comparisons showed that our model was perceived as more realistic and consistent with social behaviors compared to traditional collision avoidance approaches, that used only locomotion or random animations. The immersive user study showed that participants effectively detected the social behaviours included in our model as intended. The results revealed significant differences in the participants’ perceptions of the various behaviours exhibited by our social crowd model., This work was supported by the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No. 860768 (CLIPE project), and by MCIN/AEI/10.13039/501100011033/FEDER, UE (PID2021-122136OBC21)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

Stretch your reach: studying self-avatar and controller misalignment in virtual reality interaction

  • Pontón Martínez, José Luis|||0000-0001-6576-4528
  • Keshavarz, Reza
  • Beacco Porres, Alejandro|||0000-0001-8192-1431
  • Pelechano Gómez, Núria|||0000-0002-1437-245X
Immersive Virtual Reality typically requires a head-mounted display (HMD) to visualize the environment and hand-held controllers to interact with the virtual objects. Recently, many applications display full-body avatars to represent the user and animate the arms to follow the controllers. Embodiment is higher when the self-avatar movements align correctly with the user. However, having a full-body self-avatar following the user’s movements can be challenging due to the disparities between the virtual body and the user’s body. This can lead to misalignments in the hand position that can be noticeable when interacting with virtual objects. In this work, we propose five different interaction modes to allow the user to interact with virtual objects despite the self-avatar and controller misalignment and study their influence on embodiment, proprioception, preference, and task performance. We modify aspects such as whether the virtual controllers are rendered, whether controllers are rendered in their real physical location or attached to the user’s hand, and whether stretching the avatar arms to always reach the real controllers. We evaluate the interaction modes both quantitatively (performance metrics) and qualitatively (embodiment, proprioception, and user preference questionnaires). Our results show that the stretching arms solution, which provides body continuity and guarantees that the virtual hands or controllers are in the correct location, offers the best results in embodiment, user preference, proprioception, and performance. Also, rendering the controller does not have an effect on either embodiment or user preference., This work has received funding from the European Union’sHorizon 2020 research and innovation programme under HORIZON-CL42022-HUMAN-01 grant agreement No 101093159 (XR4ED), and from MCIN/AEI/10.13039/501100011033/FEDER "A way to make Europe", UE (PID2021-122136OB-C21). Jose Luis Ponton was also funded by the Spanish Ministry of Universities (FPU21/01927)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

A 3D feature-based approach for mapping scaling effects on stone monuments

  • Muñoz Pandiella, Imanol|||0000-0002-4781-7574
  • Pueyo Sandez, Francesc Xavier|||0000-0002-3622-583X
  • Bosch Geli, Carles
Weathering effects caused by physical, chemical, or biological processes result in visible damages that alter the appearance of stones’ surfaces. Consequently, weathered stone monuments can offer a distorted perception of the artworks to the point of making their interpretation misleading. Being able to detect and monitor decay is crucial for restorers and curators to perform important tasks such as identifying missing parts, assessing the preservation state, or evaluating curating strategies. Decay mapping, the process of identifying weathered zones of artworks, is essential for preservation and research projects. This is usually carried out by marking the affected parts of the monument on a 2D drawing or picture of it. One of the main problems of this methodology is that it is manual work based only on experts’ observations. This makes the process slow and often results in disparities between the mappings of the same monument made by different experts. In this paper, we focus on the weathering effect known as “scaling”, following the ICOMOS ISCS definition. We present a novel technique for detecting, segmenting, and classifying these effects on stone monuments. Our method is user-friendly, requiring minimal user input. By analyzing 3D reconstructed data considering geometry and appearance, the method identifies scaling features and segments weathered regions, classifying them by scaling subtype. It shows improvements over previous approaches and is well-received by experts, representing a significant step towards objective stone decay mapping., This work has been partially supported by projects PID2021-122136OB-C21 and PID2021-122136OB-C22 funded by MCIN/AEI/10.13039/501100011033 and by ERDF A way of making Europe and developed in the context of Grup de Recerca Consolidat ViRVIG 2021-SGR-01035 with the support of the Departament of Research and Universities of the Generalitat de Catalunya., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Digital reconstruction of partially lost altarpieces. The case of the Rosary's altarpiece of Sant Pere Màrtir de Manresa

  • Pagès Vilà, Anna
  • Pueyo Sandez, Francesc Xavier|||0000-0002-3622-583X
  • Muñoz Pandiella, Imanol|||0000-0002-4781-7574
© 2023 Author's | Association for Computing Machinery. This is the author's version of the work. It is posted here for your personal use. Not for redistribution. The definitive Version of Record was published in Journal on Computing and Cultural Heritage, http://dx.doi.org/10.1145/3652860., An important challenge of Digital Cultural Heritage is to contribute to the recovery of artworks with their original shape and appearance. Many altarpieces, which are very relevant Christian art elements, have been damaged and/or partly or fully lost. Therefore, the only way to recover them is to carry out their digital reconstruction. Although the procedure that we present here is valid for any altarpiece with similar characteristics, and even for other akin elements, our test bench is the altarpieces that were damaged or destroyed or that disappeared during the Spanish Civil War (1936–1939) in Catalonia, where most suffered these effects. The first step of our work has been the classification of these artworks into different categories on the basis of their degree of destruction and of the available visual information related to each one.</jats:p> <jats:p>This article proposes, for the first time to our knowledge, a workflow for the virtual reconstruction, through photogrammetry, digital modeling, and digital color restoration, of whole altarpieces partially preserved with very little visual information. Our case study is the Rosary’s altarpiece of Sant Pere Màrtir de Manresa church. Currently, this altarpiece is partially preserved in fragments in the Museu Comarcal de Manresa (Spain). But it cannot be reassembled physically owing to the lack of space (actually the church does not exist anymore) and the cost of such an operation. Thus, there is no other solution than the digital one to contemplate and study the altarpiece as a whole. The reconstruction that we provide allows art historians and the general public to virtually see the altarpiece complete and assembled as it was until 1936. The results obtained also allow us to see in detail the reliefs and ornaments of the altarpiece with their digitally restored color., The authors would like to thank Dr. Joan Bosch for his support and advice during the development of the research presented here and the Museu de Manresa - Museu del Barroc de Catalunya for granting access to the artworks of the Rosary’s altarpiece and permitting their digitization, and authorizing us their use. We also want to express our gratitude to the GIRAVOLT program (https://patrimoni.gencat.cat/en/giravolt) of the Departament de Cultura de la Generalitat de Catalunya for providing us with the photogrammetric models of the scenes of the Annunciation and Pentecost of the altarpiece and granting us their use. This work is part of the doctoral thesis of the first author, which is funded by the IFUdG2022-2024 grant, of the University of Girona with the collaboration of the Santander Bank. It has been partially supported by the projects PID2021-122136OB-C21 and PID2021-122136OB-C22 funded by MCIN/AEI/ 10.13039/501100011033 and by ERDF A way of making Europe and by the project PID2020-114339GB-C21 funded by the MCIN/AEI/, and developed in the context of Grup de Recerca Consolidat ViRVIG 2021-SGR-01035 with the support of the Departament of Research and Universities of the Generalitat de Catalunya., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Automatic colon segmentation on T1-FS MR images

  • Orellana Bech, Bernat
  • Navazo Álvaro, Isabel|||0000-0001-6298-1463
  • Brunet Crosa, Pere|||0000-0001-8406-1975
  • Monclús Lahoya, Eva|||0000-0002-9645-0510
  • Bendezú García, Álvaro
  • Azpiroz Vidaur, Fernando
The volume and distribution of the colonic contents provides valuable insights into the effects of diet on gut microbiotica involving both clinical diagnosis and research. In terms of Magnetic Resonance Imaging modalities, T2-weighted images allow the segmentation of the colon lumen, while fecal and gas contents can be only distinguished on the T1-weighted Fat-Sat modality. However, the manual segmentation of T1-weighted Fat-Sat is challenging, and no automatic segmentation methods are known. This paper proposed a non-supervised algorithm providing an accurate T1-weighted Fat-Sat colon segmentation via the registration of an existing colon segmentation in T2-weighted modality. The algorithm consists of two phases. It starts with a registration process based on a classical deformable registration method, followed by a novel Iterative Colon Registration process that utilizes a mesh deformation approach. This approach is guided by a probabilistic model that provides the likelihood of the colon boundary, followed by a shape preservation process of the colon segmentation on T2-weighted images. The iterative process converges to achieve an optimal fit for colon segmentation in T1-weighted Fat-Sat images. The segmentation algorithm has been tested on multiple datasets (154 scans) and acquisition machines (3) as part of the proof of concept for the proposed methodology. The quantitative evaluation was based on two metrics: the percentage of ground truth labeled feces correctly identified by our proposal (93+-5% ), and the volume variation between the existing colon segmentation in the T2-weighted modality and the colon segmentation computed in T1-weighted Fat-Sat images. Quantitative and medical evaluations demonstrated a degree of accuracy, usability, and stability concerning the acquisition hardware, making the algorithm suitable for clinical application and research., This work was supported in part by the projects PID2021-122295OB-I00 (Ministerio de Ciencia e Innovación, Spain), the project PID2021-122136OB-C21 funded by MCIN/AEI/ 10.13039/501100011033 and ERDF “A way of making Europe”, by the EU Horizon 2020 and the Department of Research and Universities of the Government of Catalonia (2021 SGR 01035). Ciberehd is funded by the Instituto de Salud Carlos III, Spain., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Simulating real-life scenarios to better understand the spread of diseases under different contexts

  • Blanco Guerra, Rafael Andrés
  • Patow, Gustavo Ariel|||0000-0002-1977-9101
  • Pelechano Gómez, Núria|||0000-0002-1437-245X
Current statistical models to simulate pandemics miss the most relevant information about the close atomic interactions between individuals which is the key aspect of virus spread. Thus, they lack a proper visualization of such interactions and their impact on virus spread. In the field of computer graphics, and more specifically in computer animation, there have been many crowd simulation models to populate virtual environments. However, the focus has typically been to simulate reasonable paths between random or semi-random locations in a map, without any possibility of analyzing specific individual behavior. We propose a crowd simulation framework to accurately simulate the interactions in a city environment at the individual level, with the purpose of recording and analyzing the spread of human diseases. By simulating the whereabouts of agents throughout the day by mimicking the actual activities of a population in their daily routines, we can accurately predict the location and duration of interactions between individuals, thus having a model that can reproduce the spread of the virus due to human-to-human contact. Our results show the potential of our framework to closely simulate the virus spread based on real agent-to-agent contacts. We believe that this could become a powerful tool for policymakers to make informed decisions in future pandemics and to better communicate the impact of such decisions to the general public., This work has received funding from the European Union’s Horizon 2020 research and innovation program under the Marie Skłodowska Curie grant agreement No 860768 (CLIPE project). Also, it was partially funded by grants PID2021-122136OB- C21 and PID2021-122136OB-C22 funded by MCIN/AEI/https://doi.org/10.13039/ 50110 0011033 and by ERDF A way of making Europe., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Digital 3D models for medieval heritage: diachronic analysis and documentation of its architecture and paintings

  • Muñoz Pandiella, Imanol|||0000-0002-4781-7574
  • Bosch Geli, Carles
  • Guardia Pons, Milagros
  • Cayuela Vellido, Begoña
  • Pogliani, Paola
  • Bordi, Giulia
  • Paschali, Maria
  • Andújar Gran, Carlos Antonio|||0000-0002-8480-4713
  • Charalambous, Panayiotis
In this paper, we discuss the requirements and technical challenges within the EHEM project, Enhancement of Heritage Experiences: The Middle Ages, an ongoing research program for the acquisition, analysis, documentation, interpretation, digital restoration, and communication of medieval artistic heritage. The project involves multidisciplinary teams comprising art historians and visual computing experts. Despite the vast literature on digital 3D models in support of Cultural Heritage, the field is so rich and diverse that specific projects often imply distinct, unique requirements which often challenge the computational technologies and suggest new research opportunities. As good representatives of such diversity, we describe the three monuments that serve as test cases for the project, all of them with a rich history of architecture and paintings. We discuss the art historians’ view of how digital models can support their research, the expertise and technological solutions adopted so far, as well as the technical challenges in multiple areas spanning geometry and appearance acquisition, color analysis and digital restitution, as well as the representation of the profound transformations due to the alterations suffered over the centuries., The EHEM project was approved in the JPI Cultural Heritage Conservation, Protection, and Use (0127) call in 2020. The JPICH has received funding from the European Union’s Horizon 2020 research and innovation program under grant agreement no. 699523. This project has received funding from the European Union’s Horizon 2020 Research and Innovation Programme under Grant Agreement No. 739578 and the Government of the Republic of Cyprus through the Deputy Ministry of Research, Innovation, and Digital Policy. This work has been partially funded from MCIN/AEI/10.13039/501100011033/FEDER, UE (Spain) in the framework of the project PID2021-122136OB-C21, and with the support of the Department of Research and Universities of the Government of Catalonia (2021 SGR 01035)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Visualizing game dynamics at a specific time: Influence of the players’ poses for tactical analyses in padel

  • Javadiha, Mohammadreza|||0000-0002-4867-1132
  • Andújar Gran, Carlos Antonio|||0000-0002-8480-4713
  • Lacasa Claver, Enrique
  • Shirato, Gota
  • Andrienko, Natalia
  • Andrienko, Gennady
Tactical elements are crucial in team sports. The analysis of hypothetical game situations greatly benefits from positional diagrams showing where the players are. These diagrams often show the layout of the players through simple symbols, which provide no information about their poses. This paper investigates if the visualization of player poses is benefitial for tactical understanding of positional diagrams in padel. We propose a realistic, cartoon-like representation of the players and discuss its integration into a typical positional diagram. To overcome the cost of generating player representations depicting their pose, we propose a method to generate such representations from minimal user input. We conducted a user study to evaluate the effectiveness of our pose-aware diagrams. The tasks for the study were designed to encompass the main in-game scenarios in padel, which include the ballholder at the net with opponents defending, the reverse situation, and transitions between these two states. We found that our representation is preferred over a symbolic one that only indicates player orientation. The proposed method enables coaches to produce such representations within a matter of seconds, thereby significantly facilitating the creation of detailed and easily analyzable depictions of game situations., This work was supported by MICIU/AEI/10.13039/50110001 1033 and ERDF/EU under Grant PID2021-122136OB-C21, and by the Federal Ministry of Education and Research of Germany and the state of North-Rhine Westphalia as part of the Lamarr Institute for Machine Learning and Artificial Intelligence (Lamarr22B). Mohammadreza Javadiha was also supported by the Spanish Ministry of Science and Innovation and ERDF/EU under Grant PRE2018-086835. The ViRVIG group also received funding from the Department of Research and Universities of the Government of Catalonia (2021 SGR 01035)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Exploratory analysis of scientific publications for university governance

  • Gràcia Calvo, Alexandre|||0009-0009-6728-6713
  • Padró, Lluís|||0000-0003-4738-5019
  • Alarcón Cot, Eduardo José|||0000-0001-7663-7153
  • Vázquez Alcocer, Pere Pau|||0000-0003-4638-4065
Research-oriented universities often comprise numerous researchers of various types and possess complex research structures that encompass research groups, departments, laboratories, and research institutes. In this situation, understanding the university's strengths and areas of excellence requires careful examination. Additionally, individuals at different levels of governance (e.g., department heads, directors of research institutes, rectors) may seek to establish synergies among researchers to tackle issues such as international project applications or industry technology transfer. University officials and faculty members frequently require the expertise of specific research groups or individuals, but struggle to obtain this information beyond their personal networks. This limits their ability to locate necessary resources effectively. Fortunately, most institutions have databases containing publications that could provide valuable insights into areas of strength within the university. In this article, we present a visual analysis application capable of addressing these questions and assisting management in making informed decisions regarding governance measures such as creating new research institutes. Our system has been evaluated by domain experts, who found it highly beneficial and expressed interest in utilising it regularly., This project has been supported by PID2021-122136OB-C21 from the Ministerio de Ciencia e Innovación, Spain, by 839 FEDER (EU) funds., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

PHA: Part-wise heterogeneous agents with reusable policy priors for physics-based motion synthesis

  • Carranza Cobeñas, Luis Alberto|||0009-0008-8545-148X
  • Argudo Medrano, Óscar|||0000-0003-3943-1839
  • Andújar Gran, Carlos Antonio|||0000-0002-8480-4713
Performing everyday tasks requires both large-scale body movements driven by the arms, legs, and torso, and fine motor skills, particularly in the hands. However, existing reinforcement learning approaches often struggle to efficiently acquire these diverse motion skills from scratch, leading to slow convergence or suboptimal policies. We observed that the motor skills of distinct body parts exhibit a certain level of independence. This suggests the potential advantage of independently pretraining specific dexterous limbs prior to their integration in full-body motion tasks. Inspired by this, we present Part-wise Heterogeneous Agents (PHA), a cooperative multi-agent reinforcement learning approach where body parts are treated as independent agents, allowing for specialized skill acquisition and cooperative execution of complex full-body tasks. Furthermore, our method enables the pretraining of fine motor skills, such as gripping a bar or grabbing a climbing hold, before integrating them with other body parts for complex whole-body coordination, thus introducing part-wise Reusable Policy Priors. We tested our technique on challenging tasks such as rope climbing, rock bouldering and traversing a horizontal ladder. Our approach not only accelerates convergence, but also improves overall policy quality, achieving motion tasks that single-agent approaches struggle to solve. Our results also demonstrate adaptability, enabling Reusable Policy Priors to adjust their policies to successfully perform complex tasks in scenarios not seen during training., This work was supported by Grant PID2021-122136OB-C21, funded by MICIU/AEI/10.13039/501100011033 and ERDF/EU, as well as the Department of Research and Universities of the Government of Catalonia (2021 SGR 01035). Luis Carranza was also supported with the grant PRE2022-102912 funded by MICIU/AEI/10.13039/501100011033 and ESF+., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

Evaluating the suitability of long document embeddings for classification tasks: a comparative analysis

  • Rafieian, Bardia
  • Vázquez Alcocer, Pere Pau|||0000-0003-4638-4065
Long documents pose a significant challenge for natural language processing (NLP), which requires high-quality embeddings. Despite the numerous approaches that encompass both deep learning and machine learning methodologies, tackling this task remains hard. In our study, we tackle the issue of long document classification by leveraging recent advancements in machine learning and deep learning. We conduct a comprehensive evaluation of several state-of-the-art models, including Doc2vec, Longformer, LLaMA-3, and SciBERT, focusing on their effectiveness in handling long to very long documents (in number of tokens). Furthermore, we trained a Doc2vec model using a massive dataset, achieving state-of-the-art quality, and surpassing other methods such as Longformer and SciBERT, which are very costly to train. Notably, while LLaMA-3 outperforms our model in certain aspects, Doc2vec remains highly competitive, particularly in speed, as it is the fastest among the evaluated methods. Through exp erimentation, we thoroughly evaluate the performance of our custom-trained Doc2vec model in classifying documents with an extensive number of tokens, demonstrating its efficacy, especially in handling very long documents. However, our analysis also uncovers inconsistencies in the performance of all models when faced with documents containing larger text volumes., This project has been supported by PID2021-122136OB-C21 from the Ministerio de Ciencia e Innovacion, by 839 FEDER (EU) funds., Peer Reviewed, Objectius de Desenvolupament Sostenible::9 - Indústria, Innovació i Infraestructura
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Evaluating graphical perception capabilities of Vision Transformers

  • Poonam, Poonam|||0009-0002-0472-229X
  • Vázquez Alcocer, Pere Pau|||0000-0003-4638-4065
  • Ropinski, Timo
Vision Transformers (ViTs) have emerged as a powerful alternative to convolutional neural networks (CNNs) in a variety of image-based tasks. While CNNs have previously been evaluated for their ability to perform graphical perception tasks, which are essential for interpreting visualizations, the perceptual capabilities of ViTs remain largely unexplored. In this work, we investigate the performance of ViTs in elementary visual judgment tasks inspired by Cleveland and McGill’s foundational studies, which quantified the accuracy of human perception across different visual encodings. Inspired by their study, we benchmark ViTs against CNNs and human participants in a series of controlled graphical perception tasks. Our results reveal that, although ViTs demonstrate strong performance in general vision tasks, their alignment with human-like graphical perception in the visualization domain is limited. This study highlights key perceptual gaps and points to important considerations for the application of ViTs in visualization systems and graphical perceptual modeling., Pere-Pau Vázquez was supported by grant PID2021-122136OB-C21 from the Ministerio de Ciencia e Innovación (Spain), by FEDER (EU) funds (project 839), and by 2021 SGR 01035 from the Generalitat de Catalunya., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Virtual reality inspection of chromatin 3D and 2D data

  • Molina López, Elena|||0000-0003-4561-6215
  • Kouril, David
  • Isenberg, Tobias
  • Kozliková, Barbora
  • Vázquez Alcocer, Pere Pau|||0000-0003-4638-4065
Understanding the packing of long DNA strands into chromatin is one of the ultimate challenges in genomic research. An intrinsic part of this complex problem is studying the chromatin’s spatial structure. Biologists reconstruct 3D models of chromatin from experimental data, yet the exploration and analysis of such 3D structures is limited in existing genomic data visualization tools. To improve this situation, we investigated the current options of immersive methods and designed a prototypical VR visualization tool for 3D chromatin models that leverages virtual reality to deal with the spatial data. We showcase the tool in three primary use cases. First, we provide an overall 3D shape overview of the chromatin to facilitate the identification of regions of interest and the selection for further investigation. Second, we include the option to export the selected regions and elements in the BED format, which can be loaded into common analytical tools. Third, we integrate epigenetic modification data along the sequence that influence gene expression, either as in-world 2D charts or overlaid on the 3D structure itself. We developed our application in collaboration with two domain experts and gathered insights from two informal studies with five other experts., This project has been supported by PID2021-122136OB-C21 from the Ministerio de Ciencia e Innovación, by 839 FEDER (EU) funds, and The Czech Science Foundation research project no. GA23-05651S. Elena Molina has been supported by an FPU grant from the Ministerio de Ciencia e Innovación., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Seeing through others’ eyes in immersive VR: insights from a user study

  • Monclús Lahoya, Eva|||0000-0002-9645-0510
  • Fairén González, Marta|||0000-0001-7293-584X
  • Navazo Álvaro, Isabel|||0000-0001-6298-1463
  • Vázquez Alcocer, Pere Pau|||0000-0003-4638-4065
Despite the increasing usage of collaborative Virtual Reality in classrooms, teachers often struggle to ascertain whether the participants are adequately sharing the same viewpoint and are seeing the same information during lectures. Virtual Reality offers numerous advantages, including immersive experiences that enhance traditional classroom instruction and foster active and cooperative learning. A relevant part of the teaching process involves pointing to concrete regions or points of the virtual scene, and describing them. At this point, it is important that the pupils are able to see the same that is being seen by the instructor. In this paper we analyze viewpoint sharing methods commonly used in the literature, like teleport, 2D views with or without continuous streaming, and propose a novel method, dubbed Model Movement (MM). Our application layout is structured around a central 3D model that acts as the primary focal point at the collaborative session and the experiments were performed using 3D anatomical models. We conducted a user study with 32 participants (local and nominally remote) where participants were asked multiple times to swap to the instructor’s viewpoint and tracked their response times and their movement in the scene, among other variables. The data analysis and results of the questionnaires indicate that MM appears to be the preferred, more effective and usable of the analyzed techniques at showing the region of interest to inspect., This work has been partially supported by the project PID2021-122136OB-C21 funded by MCIN/AEI/ 10.13039/501100011033 and ERDF “A way of making Europe”, by the EU Horizon 2020 and the Department of Research and Universities of the Government of Catalonia (2021 SGR 01035), Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

Reconstructing lost altarpieces: a differentiable rendering approach

  • Pagès Vilà, Anna
  • Muñoz Pandiella, Imanol|||0000-0002-4781-7574
Studying works that have completely or partially disappeared is always difficult due to the lack of information. In more fortunate scenarios where photographs were taken before the destruction, the study of the piece is limited by the viewpoints captured in the available photographs. In this interdisciplinary research, we present a new methodology for reconstructing lost altarpieces from a single historical image, utilizing differentiable rendering techniques. We test our methodology by reconstructing some reliefs from the altarpiece of Sant Joan Baptista (Valls, Spain), which was destroyed in 1936. These results are valuable for both experts and the public, as they facilitate a better understanding of the relief's volumetrics and their spatial relationships, representing a significant advancement in the virtual recovery of lost artifacts., This work has been partially supported by the projects PID2021-122136OB-C21andPID2021-122136OB-C22fundedby MCIN/AEI/ 10.13039/501100011033 and by FEDER A way of making Europe, by the project PID2020-114339GB-C21 funded by the MCIN/AEI, and by the Department of Research and Universities of the Generalitat de Catalunya (2021 SGR 01035)., This work has been partially supported by the projects PID2021-122136OB-C21 and PID2021-122136OB-C22 funded by MCIN/AEI/ 10.13039/501100011033 and by FEDER A way of making Europe, by the project PID2020-114339GB-C21 funded by the MCIN/AEI, and by the Department of Research and Universities of the Generalitat de Catalunya (2021 SGR 01035)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

TreeON: Reconstructing 3D tree point clouds from orthophotos and heightmaps

  • Grammatikaki, Angeliki
  • Eschner, Johannes
  • Hermosilla Casajús, Pedro
  • Argudo Medrano, Óscar|||0000-0003-3943-1839
  • Waldner, Manuela
We present TreeON, a novel neural-based framework for reconstructing detailed 3D tree point clouds from sparse top-down geodata, using only a single orthophoto and its corresponding Digital Surface Model (DSM). Our method introduces a new training supervision strategy that combines both geometric supervision and a differentiable shadow and silhouette loss to learn point cloud representations of trees without requiring species labels, procedural rules, detailed terrestrial reconstruction data, or ground laser scan data. To address the lack of ground truth data, we generate a synthetic dataset of point clouds from procedurally modeled trees and train our network on it. Quantitative and qualitative experiments demonstrate better reconstruction quality and coverage compared to existing methods, as well as strong generalization to real-world data, leading to visually appealing and structurally plausible tree point cloud representations that can be integrated into interactive digital 3D maps. The codebase, synthetic dataset, and pretrained model are publicly available at https://angelikigram.github.io/treeON/., The authors acknowledge TU Wien Bibliothek for financial support through its Open Access Funding Programme. The VRVisGmbH is funded by BMIMI, BMWET, Tyrol, Vorarlberg and Vienna Business Agency in the scope of COMET - Competence Centers for Excellent Technologies (904918, 911654) which is managed by FFG. This research has been funded by WWTF project ICT22-055 - Instant Visualization and Interaction for Large Point Clouds, and Grant PID2021-122136OB-C21 funded by MI-CIU/AEI/10.13039/501100011033 and ERDF/EU., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

DragPoser: motion reconstruction from variable sparse tracking signals via latent space optimization

  • Pontón Martínez, José Luis|||0000-0001-6576-4528
  • Pujol Puig, Eduard|||0000-0001-6406-7499
  • Aristidou, Andreas
  • Andújar Gran, Carlos Antonio|||0000-0002-8480-4713
  • Pelechano Gómez, Núria|||0000-0002-1437-245X
High-quality motion reconstruction that follows the user's movements can be achieved by high-end mocap systems with many sensors. However, obtaining such animation quality with fewer input devices is gaining popularity as it brings mocap closer to the general public. The main challenges include the loss of end-effector accuracy in learning-based approaches, or the lack of naturalness and smoothness in IK-based solutions. In addition, such systems are often finely tuned to a specific number of trackers and are highly sensitive to missing data, e.g., in scenarios where a sensor is occluded or malfunctions. In response to these challenges, we introduce DragPoser, a novel deep-learning-based motion reconstruction system that accurately represents hard and dynamic constraints, attaining real-time high end-effectors position accuracy. This is achieved through a pose optimization process within a structured latent space. Our system requires only one-time training on a large human motion dataset, and then constraints can be dynamically defined as losses, while the pose is iteratively refined by computing the gradients of these losses within the latent space. To further enhance our approach, we incorporate a Temporal Predictor network, which employs a Transformer architecture to directly encode temporality within the latent space. This network ensures the pose optimization is confined to the manifold of valid poses and also leverages past pose data to predict temporally coherent poses. Results demonstrate that DragPoser surpasses both IK-based and the latest data-driven methods in achieving precise end-effector positioning, while it produces natural poses and temporally coherent motion. In addition, our system showcases robustness against on-the-fly constraint modifications, and exhibits adaptability to various input configurations and changes. The complete source code, trained model, animation databases, and supplementary material used in this paper can be found at https://upc-virvig.github.io/DragPoser, This work has received funding from the European Union’s Horizon 2020 Research and Innovation Programme under HORIZON-CL4-2022-HUMAN-01 grant agreement No 101093159 (XR4ED) and grant agreement No 739578 (RISE), from MCIN/AEI/10.13039/501100011033/FEDER “A way to make Europe”, UE (PID2021-122136OB-C21), from the Departament of Research and Universities of the Government of Catalonia (2021 SGR 01035), and the Government of the Republic of Cyprus through the Deputy Ministry of Research, Innovation and Digital Policy. Jose Luis Ponton was also funded by the Spanish Ministry of Universities (FPU21/01927), and Eduard Pujol by Universitat Politècnica de Catalunya and Banco Santander (FPI-UPC predoctoral grant)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Monkey see, monkey break? Study of rule-breaking imitation in virtual crowds

  • Huang, Kexiang
  • Yin, Tairan|||0000-0001-6023-6507
  • Pontón Martínez, José Luis|||0000-0001-6576-4528
  • Tang, Ruida
  • Itatani, Reiya|||0000-0002-8430-1657
  • Ding, Gangyi
  • Argudo Medrano, Óscar|||0000-0003-3943-1839
  • Pelechano Gómez, Núria|||0000-0002-1437-245X
Rule-breaking behaviors, such as jaywalking or skipping queues, are common in crowds but difficult to study in real-world settings due to limited control and observability. Virtual reality (VR) provides a controlled alternative, but its validity depends on whether VRelicits realistic rule-breaking behavior. We conducted a VR study with 65 participants navigating a virtual city with four scenarios differing in social norms: walking on grass, crossing outside a crosswalk, jaywalking at a red light, and skipping a line. In each scenario, the proportion of rule-breaking virtual characters was manipulated (0%, 25%, 50%). Participants' movements and gaze were recorded to assess behavior and attention. Results showed higher rule-breaking as the number of violators increased, except in the low-stakes crossing scenario. Rule-breakers attended more to violating characters, whereas rule-followers focused on compliant ones. Participants cited efficiency, safety, and social norms as key factors guiding their decisions. Overall, VR reproduced natural patterns of social compliance and noncompliance, supporting its use for studying rule-breaking and applications in crowd simulation, urban design, safety training, and immersive media., This work was supported by grants PID2021-122136OB-C21 from MCIN/AEI/10.13039/501100011033/FEDER “A way to make Europe” and the National Natural Science Foundation of China (No.62177005), Peer Reviewed, 11 - Ciutats i Comunitats Sostenibles
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Towards the virtual reconstruction of physically lost altarpieces through differentiable rendering

  • Pagès Vilà, Anna
  • Muñoz Pandiella, Imanol|||0000-0002-4781-7574
The interest of human beings in artistic creation, and its preservation, has been a constant throughout history. Thanks to this, today we can enjoy works from various moments and origins. However, not all works have been so fortunate, and many have disappeared for various reasons. To mitigate this loss, the virtual recreation of lost works of art is an open topic of interest for the computer graphics and cultural heritage communities. In the literature, we can find works along this line that are based on manual modeling or the use of numerous modern photographs. In this interdisciplinary work, however, we present a new methodology for reconstructing destroyed altarpieces for which only a single historical image is preserved, leveraging modern techniques of differentiable rendering. Additionally, we test our methodology in the reconstruction of the altarpiece of Sant Joan Baptista (Valls, Spain) destroyed during the Spanish Civil War. Our approach results in the partial 3D reconstruction of the main reliefs of the altarpiece. These results are truly useful, both for experts and for the public, as they allow for a better understanding of its volumetrics and its relationship with space, representing an important step forward in the virtual recovery of lost artifacts., This work has been partially supported by the projects PID2021-122136OB-C21 and PID2021-122136OB-C22 funded by MCIN/AEI/ 10.13039/501100011033 and by FEDER A way of making Europe, by the project PID2020- 114339GB-C21 funded by the MCIN/AEI, and by the Department of Research and Universities of the Generalitat de Catalunya (2021 SGR 01035)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

AI-in-the-loop: The future of biomedical visual analytics applications in the era of AI

  • Bühler, Katja
  • Höllt, Thomas
  • Schultz, Thomas
  • Vázquez Alcocer, Pere Pau|||0000-0003-4638-4065
AI is the workhorse of modern data analytics and omnipresent across many sectors. Large language models and multimodal foundation models are today capable of generating code, charts, visualizations, etc. How will these massive developments of AI in data analytics shape future data visualizations and visual analytics workflows? What is the potential of AI to reshape methodology and design of future visual analytics applications? What will be our role as visualization researchers in the future? What are opportunities, open challenges, and threats in the context of an increasingly powerful AI? This Visualization Viewpoints discusses these questions in the special context of biomedical data analytics as an example of a domain in which critical decisions are taken based on complex and sensitive data, with high requirements on transparency, efficiency, and reliability. We map recent trends and developments in AI on the elements of interactive visualization and visual analytics workflows and highlight the potential of AI to transform biomedical visualization as a research field. Given that agency and responsibility have to remain with human experts, we argue that it is helpful to keep the focus on human-centered workflows, and to use visual analytics as a tool for integrating “AI-in-the-loop.” This is in contrast to the more traditional term “human-in-the-loop.” which focuses on incorporating human expertise into AI-based systems., The work of Katja Buhler was supported by VRVis through BMK, BMAW, Styria, SFG, Tyrol, and Vienna Business Agency in the scope of COMET—Competence Centers for Excellent Technologies managed by FFG under Grant 879730. The work of Thomas Hollt was supported by the Sector Plan Beta en Techniek at TU Delft. The work of Pere-Pau Vazquez was supported by the Ministerio de Ciencia e Innovacion, Spain, through 839 FEDER (EU) funds under Grant PID2021-122136OB-C21, and by Generalitat de Catalunya, 2021 SGR 01035. This work has benefited from Dagstuhl Seminar 23451 “Visualization of Biomedical Data – Shaping the Future and Building Bridges on Biomedical Data., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Terrain synthesis and authoring based on iso-contours

  • Huftier, Benoit|||0009-0002-4726-9392
  • Schott, Hugo
  • Galin, Eric
  • Argudo Medrano, Óscar|||0000-0003-3943-1839
  • Peytavie, Adrien
  • Guerin, Eric
Digital terrains are central to realistic landscape depiction, yet authoring tools must balance perceptual realism with intuitive artistic control. We propose a compact vector-based representation that models terrain as nested iso-contours, inspired by geomorphology and cartography. Our method departs from traditional grid-based elevation models by generating contours through an inward Open Eden Growth simulation, followed by marching-triangles reconstruction into a Triangulated Irregular Network. This contour framework supports direct editing such as warping, slope modulation, and smoothing, while allowing reconstruction of a standard elevation map for downstream processing, including erosion and amplification. The approach enables the creation of diverse, realistic terrains from minimal user input and offers simple yet powerful control for designers., This work is funded by the project EOLE ANR-23-CE56-0008, supported by Agence Nationale de la Recherche (France), and Grant PID2021-122136OB-C21 funded by MI-CIU/AEI/10.13039/501100011033 and ERDF/EU. Open access publication funding provided by COUPERIN CY26., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

DRL-driven padel players: Simulating padel matches through deep reinforcement learning in real and hypothetical scenarios

  • Javadiha, Mohammadreza|||0000-0002-4867-1132
  • Ji, Jia Long
  • Zhou, Wenqi
  • Lacasa Claver, Enrique
  • Andújar Gran, Carlos Antonio|||0000-0002-8480-4713
Recent advances in Deep Reinforcement Learning (DRL) have opened new avenues for sport research. DRL allows virtual agents to learn and solve complex tasks with minimal input, which means that models can be trained with little or no data collection. This enables the creation of sport simulations that go beyond the replication of real matches, exploring entirely new conditions where prior knowledge is limited. In this paper, we investigate the generation of simulated data for padel matches, aiming to either replicate real environments or investigate hypothetical ones with altered parameters such as non-standard court sizes. We focus on high-level behavioural aspects of the players rather than low-level skills. We present a proof-of-concept implementation of a DRL system that enables robot-like agents to learn and play padel under varying conditions. After training different types of virtual agents, we observed their overall behaviour during simulated matches. In particular, for certain combinations of parameters, the agents learned to coordinate their movements similarly to professional players. We argue that DRL-driven players have the potential to complement traditional mathematical models, especially in scenarios where data collection is challenging or unfeasible. This could lead to more engaging, safer and more inclusive sports in the future., This work was supported by the MICIU/AEI/10.13039/501100011033 and ERDF/EU under [Grant PID2021-122136OB-C21], and the Department of Research andUniversities of the Government of Catalonia [2021 SGR 01035]., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

Simulation of mechanical weathering for modeling rocky terrains

  • Mateos Arlanzón, Diego
  • Carranza Cobeñas, Luis Alberto|||0009-0008-8545-148X
  • Susín Sánchez, Antonio|||0000-0002-0874-2784
  • Argudo Medrano, Óscar|||0000-0003-3943-1839
Synthetic terrains play a vital role in various applications, including entertainment, training, and simulation. This work focuses on rocky terrains akin to those found in alpine environments, which contain many complex features such as sharp ridges, loose blocks, or overhangs that are often inadequately represented by standard 2D elevation maps. We propose a novel method based on a simplified simulation of mechanical erosion processes commonly observed in high-altitude terrains, in particular the weathering due to freeze-thaw cycles. The ultimate objective is to generate plausible rocky geometry from existing 3D models, as well as account for the temporal evolution due to these weathering processes. Additionally, we have developed an artist-friendly tool integrated as an add-on into Blender., This work has been funded with Grant PID2021-122136OB-C21 funded by MICIU/AEI/10.13039/501100011033 and ERDF/EU, and with the support of the Department of Research and Universi- ties of the Government of Catalonia (2021 SGR 01035). Luis Carranza was supported with Grant PRE2022-102912 funded by MI- CIU/AEI/10.13039/501100011033 and ESF+, and Oscar Argudo with a Maria Zambrano fellowship from Ministerio de Universidades and European Union—NextGenerationEU., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

Rediscovering mural paintings: experiencing medieval art as originally conceived through historical light simulation

  • Muñoz Pandiella, Imanol|||0000-0002-4781-7574
  • Kaur Singh, Manvir
  • Bosch Geli, Carles
  • Andújar Gran, Carlos Antonio|||0000-0002-8480-4713
  • Pueyo Sandez, Francesc Xavier|||0000-0002-3622-583X
The lighting of Cultural Heritage artifacts plays a crucial role in how we perceive and consequently understand artworks. However, lighting is typically designed to enhance the experience of contemporary visitors, often diverging significantly from the original conditions and techniques under which these works were created. This disconnect between historical and modern lighting conditions makes it difficult to fully understand the original visual experience. This issue is particularly evident in the case of Romanesque paintings, which are now exhibited in well-lit museums, but were originally displayed in dimly lit churches. In this paper, we present a method to bridge this gap, focusing on the paintings of a Romanesque church. We achieve this by simulating the original lighting conditions. Our approach encompasses the entire pipeline, from acquiring data of equivalent historical light sources and computing natural lighting to performing physically based rendering for accurate light simulation. Additionally, we have developed a web application that allows users to inspect and compare the resulting HDR images using different tone mapping and luminance operators. Our work provides a valuable tool for art historians and the general public to explore different lighting hypotheses and gain a deeper understanding of the experience of visiting a medieval church as originally conceived., This work has been partially supported by the projects PID2021-122136OB-C21 and PID2021-122136OB-C22 funded by MCIN/AEI/ 10.13039/501100011033 and by FEDER A way of making Europe, and by the Department of Research and Universities of the Generalitat de Catalunya (2021 SGR 01035)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Revealing the layers of history: enriching medieval cultural heritage 4D models with interactive narratives

  • Andújar Gran, Carlos Antonio|||0000-0002-8480-4713
  • Cayuela Vellido, Begoña
  • Lorès Otzet, Immaculada
  • Bosch Geli, Carles
  • Guardia Pons, Milagros
  • Muñoz Pandiella, Imanol|||0000-0002-4781-7574
Cultural heritage (CH) preservation is increasingly leveraging digital technologies to support art history research and provide educational experiences to the public. In the case of some Medieval monuments, the myriad of architectural and pictorial transformations suffered due to the passage of time, and the geographical dispersion of its remaining decoration substantially hinders their understanding. In this article, we present a novel approach for enriching multi-layered, multi-phase 3D models of CH monuments with art history knowledge. Such knowledge is presented in the form of image-based annotations, interactive narratives, points of interest, text, and audio, organized into a collection of guided tours enabling visitors to traverse through different historical phases of the monument. A key benefit of our approach is that art historians require no programming skills to easily define all the elements of the interactive narratives. Our approach integrates various multimedia elements seamlessly into 3D models, offering visitors an engaging and informative journey through space and time. We conducted a pilot user evaluation to demonstrate the effectiveness of our approach, highlighting its value in enhancing visitor engagement and understanding of CH. The study also provided valuable feedback on the user experience, which allowed us to enhance the final version of the web-based application and the associated narratives., This work has been partially supported by projects PID2021-122136OB-C21 and PID2021-122136OB-C22 funded by MCIN/AEI/10.13039/501100011033 and ERDF “A way of making Europe,” by the EU Horizon 2020, JPICH Conservation, Protection and Use initiative (JPICH-0127), the Spanish Agencia Estatal de Investigación (grant PCI2020-111979), and the Department of Research and Universities of the Government of Catalonia (2021 SGR 01035)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Bayesian 3D shape reconstruction from noisy points and normals

  • Pujol Puig, Eduard|||0000-0001-6406-7499
  • Chica Calaf, Antonio|||0000-0003-0270-2332
Reconstructing three-dimensional shapes from point clouds remains a central challenge in geometry processing, particularly due to the inherent uncertainties in real-world data acquisition. In this work, we introduce a novel Bayesian framework that explicitly models and propagates uncertainty from both input points and their estimated normals. Our method incorporates the uncertainty of normals derived via Principal Component Analysis (PCA) from noisy input points. Building upon the Smooth Signed Distance (SSD) reconstruction algorithm, we integrate a smoothness prior based on the curvatures of the resulting implicit function following Gaussian behavior. Our method reconstructs a shape represented as a distribution, from which sampling and statistical queries regarding the shape's properties are possible. Additionally, because of the high cost of computing the variance of the resulting distribution, we develop efficient techniques for variance computation. Our approach thus combines two common steps of the geometry processing pipeline, normal estimation and surface reconstruction, while computing the uncertainty of the output of each of these steps., This work has been partially funded by Ministeri de Ciència i Innovació (MICIN), Agencia Estatal de Investigación(AEI) and the Fons Europeu de Desenvolupament Regional (FEDER) (project PID2021-122136OB-C21 funded by MCIN/AEI/10.13039/501100011033/FEDER, UE); and by the Department of Research and Universities of the Government of Catalonia (2021 SGR 01035). The first author gratefully acknowledges the Universitat Politècnica de Catalunya and Banco Santander for the financial support of his predoctoral grant FPI-UPC grant. Open Access funding enabled and organized by Universitat Politècnica de Catalunya., Peer Reviewed
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Automatic colon segmentation on T1-FS MR images

  • Orellana Bech, Bernat
  • Navazo, Isabel
  • Brunet, Pere
  • Monclús, Eva
  • Bendezú García, Rogger Alvaro
  • Azpiroz, Fernando
Colon contents; Colon segmentation; Medical image analysis, Contingut del còlon; Segmentació del còlon; Anàlisi d'imatges mèdiques, Contenido del colon; Segmentación del colon; Análisis de imágenes médicas, The volume and distribution of the colonic contents provides valuable insights into the effects of diet on gut microbiotica involving both clinical diagnosis and research. In terms of Magnetic Resonance Imaging modalities, T2-weighted images allow the segmentation of the colon lumen, while fecal and gas contents can be only distinguished on the T1-weighted Fat-Sat modality. However, the manual segmentation of T1-weighted Fat-Sat is challenging, and no automatic segmentation methods are known. This paper proposed a non-supervised algorithm providing an accurate T1-weighted Fat-Sat colon segmentation via the registration of an existing colon segmentation in T2-weighted modality. The algorithm consists of two phases. It starts with a registration process based on a classical deformable registration method, followed by a novel Iterative Colon Registration process that utilizes a mesh deformation approach. This approach is guided by a probabilistic model that provides the likelihood of the colon boundary, followed by a shape preservation process of the colon segmentation on T2-weighted images. The iterative process converges to achieve an optimal fit for colon segmentation in T1-weighted Fat-Sat images. The segmentation algorithm has been tested on multiple datasets (154 scans) and acquisition machines (3) as part of the proof of concept for the proposed methodology. The quantitative evaluation was based on two metrics: the percentage of ground truth labeled feces correctly identified by our proposal ( ), and the volume variation between the existing colon segmentation in the T2-weighted modality and the colon segmentation computed in T1-weighted Fat-Sat images. Quantitative and medical evaluations demonstrated a degree of accuracy, usability, and stability concerning the acquisition hardware, making the algorithm suitable for clinical application and research., This work was supported in part by the projects PID2021-122295OB-I00 (Ministerio de Ciencia e Innovación, Spain), the project PID2021-122136OB-C21 funded by MCIN/AEI/ 10.13039/501100011033 and ERDF “A way of making Europe”, by the EU Horizon 2020 and the Department of Research and Universities of the Government of Catalonia (2021 SGR 01035). Ciberehd is funded by the Instituto de Salud Carlos III, Spain .
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Adaptive approximation of signed distance fields through piecewise continuous interpolation

  • Pujol Puig, Eduard
  • Chica Calaf, Antonio
In this paper, we present an adaptive structure to represent a signed distance field through trilinear or tricubic interpolation of values, and derivatives, that allows for fast querying of the field. We also provide a method to decide when to subdivide a node to achieve a provided threshold error. Both the numerical error control, and the values needed to build the interpolants, require the evaluation of the input field. Still, both are designed to minimize the total number of evaluations. C0 continuity is guaranteed for both the trilinear and tricubic version of the algorithm. Furthermore, we describe how to preserve C1 continuity between nodes of different levels when using a tricubic interpolant, and provide a proof that this property is maintained. Finally, we illustrate the usage of our approach in several applications, including direct rendering using sphere marching., This work has been partially funded by Ministeri de Ciència i Innovació (MICIN), Agencia Estatal de Investigación (AEI) and the Fons Europeu de Desenvolupament Regional (FEDER) (project PID2021-122136OB-C21 funded by MCIN/AEI/10.13039/501100011033/FEDER, UE). The first author gratefully acknowledges the Universitat Politècnica de Catalunya and Banco Santander for the financial support of his predoctoral grant FPI-UPC grant., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Evaluating graphical perception capabilities of Vision Transformers

  • Poonam, Poonam
  • Vázquez Alcocer, Pere Pau
  • Ropinski, Timo
Vision Transformers (ViTs) have emerged as a powerful alternative to convolutional neural networks (CNNs) in a variety of image-based tasks. While CNNs have previously been evaluated for their ability to perform graphical perception tasks, which are essential for interpreting visualizations, the perceptual capabilities of ViTs remain largely unexplored. In this work, we investigate the performance of ViTs in elementary visual judgment tasks inspired by Cleveland and McGill’s foundational studies, which quantified the accuracy of human perception across different visual encodings. Inspired by their study, we benchmark ViTs against CNNs and human participants in a series of controlled graphical perception tasks. Our results reveal that, although ViTs demonstrate strong performance in general vision tasks, their alignment with human-like graphical perception in the visualization domain is limited. This study highlights key perceptual gaps and points to important considerations for the application of ViTs in visualization systems and graphical perceptual modeling., Pere-Pau Vázquez was supported by grant PID2021-122136OB-C21 from the Ministerio de Ciencia e Innovación (Spain), by FEDER (EU) funds (project 839), and by 2021 SGR 01035 from the Generalitat de Catalunya., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Exploratory analysis of scientific publications for university governance

  • Gràcia Calvo, Alexandre
  • Padró, Lluís
  • Alarcón Cot, Eduardo José
  • Vázquez Alcocer, Pere Pau
Research-oriented universities often comprise numerous researchers of various types and possess complex research structures that encompass research groups, departments, laboratories, and research institutes. In this situation, understanding the university's strengths and areas of excellence requires careful examination. Additionally, individuals at different levels of governance (e.g., department heads, directors of research institutes, rectors) may seek to establish synergies among researchers to tackle issues such as international project applications or industry technology transfer. University officials and faculty members frequently require the expertise of specific research groups or individuals, but struggle to obtain this information beyond their personal networks. This limits their ability to locate necessary resources effectively. Fortunately, most institutions have databases containing publications that could provide valuable insights into areas of strength within the university. In this article, we present a visual analysis application capable of addressing these questions and assisting management in making informed decisions regarding governance measures such as creating new research institutes. Our system has been evaluated by domain experts, who found it highly beneficial and expressed interest in utilising it regularly., This project has been supported by PID2021-122136OB-C21 from the Ministerio de Ciencia e Innovación, Spain, by 839 FEDER (EU) funds., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Evaluating the suitability of long document embeddings for classification tasks: a comparative analysis

  • Rafieian, Bardia
  • Vázquez Alcocer, Pere Pau
Long documents pose a significant challenge for natural language processing (NLP), which requires high-quality embeddings. Despite the numerous approaches that encompass both deep learning and machine learning methodologies, tackling this task remains hard. In our study, we tackle the issue of long document classification by leveraging recent advancements in machine learning and deep learning. We conduct a comprehensive evaluation of several state-of-the-art models, including Doc2vec, Longformer, LLaMA-3, and SciBERT, focusing on their effectiveness in handling long to very long documents (in number of tokens). Furthermore, we trained a Doc2vec model using a massive dataset, achieving state-of-the-art quality, and surpassing other methods such as Longformer and SciBERT, which are very costly to train. Notably, while LLaMA-3 outperforms our model in certain aspects, Doc2vec remains highly competitive, particularly in speed, as it is the fastest among the evaluated methods. Through exp erimentation, we thoroughly evaluate the performance of our custom-trained Doc2vec model in classifying documents with an extensive number of tokens, demonstrating its efficacy, especially in handling very long documents. However, our analysis also uncovers inconsistencies in the performance of all models when faced with documents containing larger text volumes., This project has been supported by PID2021-122136OB-C21 from the Ministerio de Ciencia e Innovacion, by 839 FEDER (EU) funds., Peer Reviewed, Objectius de Desenvolupament Sostenible::9 - Indústria, Innovació i Infraestructura, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Effect of color palettes in heatmaps perception: a study

  • Molina López, Elena
  • Middel Soria, Carolina
  • Vázquez Alcocer, Pere Pau
Heatmaps are a widely used technique in visualization. Unfortunately, they have not been investigated in depth and little is known about the best parameterizations so that they are properly interpreted. The effect of different palettes on our ability to read values is still unknown. To address this issue, we conducted a user study, in which we analyzed the effect of two commonly used color palettes, Blues and Viridis, on value estimation and value search. As a result, we provide some suggestions for what to expect from the heatmap configurations analyzed., This paper has been supported by PID2021-122136OB-C21 from the Ministerio de Ciencia e Innovación, by 839 FEDER (EU) funds. Elena Molina has been supported by FI-SDUR doctoral grant from Generalitat de Catalunya, and FPU grant from the Ministerio de Ciencia e Innovación., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

An interaction metaphor for enhanced VR-based volume segmentation

  • Monclús Lahoya, Eva
  • Vázquez Alcocer, Pere Pau
The segmentation of medical models is a complex and time-intensive process required for both diagnosis and surgical preparation. Despite the advancements in deep learning, neural networks can only automatically segment a limited number of structures, often requiring further validation by a domain expert. In numerous instances, manual segmentation is still necessary. Virtual Reality (VR) technology can enhance the segmentation process by providing improved perception of segmentation outcomes and enabling interactive supervision by experts. But inspecting how the progress of the segmentation algorithm is evolving, and defining new seeds requires seeing the inner layers of the volume, which can be costly and difficult to achieve with typical metaphors such as clipping planes. In this paper, we introduce a wedge-shaped 3D interaction metaphor designed to facilitate VR-based segmentation through detailed inspection and guidance. User evaluations demonstrated increased satisfaction with usability and faster task completion times using the tool., Research funded by PID2021-122136OB-C21 from the Ministerio de Ciencia e Innovación, Spain, by 839 FEDER (EU) funds., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Visual analysis of multi-labelled temporal noise data from multiple sensors

  • Franco Moreno, Juan José
  • Vázquez Alcocer, Pere Pau
Environmental noise pollution is a problem for cities’ inhabitants, that can be especially severe in large cities. To implement measures that can alleviate this problem, it is necessary to understand the extent and impact of different noise sources. Although gathering data is relatively cheap, processing and analyzing the data is still complex. Besides the lack of an automatic method for labelling city sounds, maybe more important is the fact that there is not a tool that allows domain experts to analytically explore data that has been manually labelled. To solve this problem, we have created a visual analytics application that facilitates the exploration of multiple-labelled temporal data captured at four different corners of a crossing in a populated area of Barcelona, the Eixample neighborhood. Our tool consists of a series of linked interactive views that facilitate top-down (from noise events to labels) and bottom-up (from labels to time slots) exploration of the captured data., This project has been supported by grants TIN2017-88515-C2-1-R (GEN3DLIVE) from the Spanish Ministerio de Economía y Competitividad, by 839 FEDER (EU) funds, and PID2021-122136OB-C21 from MCIN/AEI/10.13039/501100011033/FEDER, EU., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Bidirectional exploration of 3D cultural heritage models and photo collections

  • Borràs Civil, Bernat
  • Muñoz Pandiella, Imanol
  • Andújar Gran, Carlos Antonio
Advances in 3D digitization have greatly improved the availability of high-quality cultural heritage datasets, often consisting of 3D models and collections of high-resolution photos. However, navigating these 3D datasets poses challenges for art history researchers, especially when they need to repeatedly locate points of interest, get close-up views of them, and analyze their spatial relationships. Existing applications for 3D model sharing often fail to provide a convenient interface to simultaneously explore the 3D model and associated photo collections. This paper proposes a web-based approach that combines a 3D view of the model with a 2D view of a group of photos. Designers can quickly arrange groups of images into 2D projections, and cultural heritage users can explore the 3D model and, simultaneously, the active group of images in the 2D view. Both views are interlinked, so that navigating the 3D model modifies the active photo group, and clicking an image in the 2D view moves the 3D camera to the 3D position the image was taken from. We tested our approach with a 3D dataset on medieval mural paintings. Our tests show that our method helps users quickly locate, inspect and relate different parts of the model, reducing navigation time and frustration. The source code is available at https://github.com/BernatBC/visualitzador-de-fotografies-en-3D., This work has been partially supported by the project PID2021-122136OBC21 funded by MCIN/AEI/ 10.13039/501100011033 and ERDF "A way of making Europe", and by the Department of Research and Universities of the Government of Catalonia (2021 SGR 01035)., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

TreeON: Reconstructing 3D tree point clouds from orthophotos and heightmaps

  • Grammatikaki, Angeliki
  • Eschner, Johannes
  • Hermosilla Casajús, Pedro
  • Argudo Medrano, Óscar
  • Waldner, Manuela
We present TreeON, a novel neural-based framework for reconstructing detailed 3D tree point clouds from sparse top-down geodata, using only a single orthophoto and its corresponding Digital Surface Model (DSM). Our method introduces a new training supervision strategy that combines both geometric supervision and a differentiable shadow and silhouette loss to learn point cloud representations of trees without requiring species labels, procedural rules, detailed terrestrial reconstruction data, or ground laser scan data. To address the lack of ground truth data, we generate a synthetic dataset of point clouds from procedurally modeled trees and train our network on it. Quantitative and qualitative experiments demonstrate better reconstruction quality and coverage compared to existing methods, as well as strong generalization to real-world data, leading to visually appealing and structurally plausible tree point cloud representations that can be integrated into interactive digital 3D maps. The codebase, synthetic dataset, and pretrained model are publicly available at https://angelikigram.github.io/treeON/., The authors acknowledge TU Wien Bibliothek for financial support through its Open Access Funding Programme. The VRVisGmbH is funded by BMIMI, BMWET, Tyrol, Vorarlberg and Vienna Business Agency in the scope of COMET - Competence Centers for Excellent Technologies (904918, 911654) which is managed by FFG. This research has been funded by WWTF project ICT22-055 - Instant Visualization and Interaction for Large Point Clouds, and Grant PID2021-122136OB-C21 funded by MI-CIU/AEI/10.13039/501100011033 and ERDF/EU., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

A Query Language for Exploratory Analysis of Video-Based Tracking Data in Padel Matches

  • Javadiha, Mohammadreza
  • Andujar, Carlos
  • Lacasa Claver, Enrique
Recent advances in sensor technologies, in particular video-based human detection, object tracking and pose estimation, have opened new possibilities for the automatic or semi-automatic per-frame annotation of sport videos. In the case of racket sports such as tennis and padel, state-of-the-art deep learning methods allow the robust detection and tracking of the players from a single video, which can be combined with ball tracking and shot recognition techniques to obtain a precise description of the play state at every frame. These data, which might include the court-space position of the players, their speeds, accelerations, shots and ball trajectories, can be exported in tabular format for further analysis. Unfortunately, the limitations of traditional table-based methods for analyzing such sport data are twofold. On the one hand, these methods cannot represent complex spatio-temporal queries in a compact, readable way, usable by sport analysts. On the other hand, traditional data visualization tools often fail to convey all the information available in the video (such as the precise body motion before, during and after the execution of a shot) and resulting plots only show a small portion of the available data. In this paper we address these two limitations by focusing on the analysis of video-based tracking data of padel matches. In particular, we propose a domain-specific query language to facilitate coaches and sport analysts to write queries in a very compact form. Additionally, we enrich the data visualization plots by linking each data item to a specific segment of the video so that analysts have full access to all the details related to the query. We demonstrate the flexibility of our system by collecting and converting into readable queries multiple tips and hypotheses on padel strategies extracted from the literature., This research was funded by the Spanish Ministry of Science and Innovation and FEDER funds, grant number PID2021-122136OB-C21, MCIN/AEI/10.13039/501100011033/FEDER, UE.
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Visualizing game dynamics at a specific time: Influence of the players’ poses for tactical analyses in padel

  • Javadiha, Mohammadreza
  • Andújar Gran, Carlos Antonio
  • Lacasa Claver, Enrique
  • Shirato, Gota
  • Andrienko, Natalia
  • Andrienko, Gennady
Tactical elements are crucial in team sports. The analysis of hypothetical game situations greatly benefits from positional diagrams showing where the players are. These diagrams often show the layout of the players through simple symbols, which provide no information about their poses. This paper investigates if the visualization of player poses is benefitial for tactical understanding of positional diagrams in padel. We propose a realistic, cartoon-like representation of the players and discuss its integration into a typical positional diagram. To overcome the cost of generating player representations depicting their pose, we propose a method to generate such representations from minimal user input. We conducted a user study to evaluate the effectiveness of our pose-aware diagrams. The tasks for the study were designed to encompass the main in-game scenarios in padel, which include the ballholder at the net with opponents defending, the reverse situation, and transitions between these two states. We found that our representation is preferred over a symbolic one that only indicates player orientation. The proposed method enables coaches to produce such representations within a matter of seconds, thereby significantly facilitating the creation of detailed and easily analyzable depictions of game situations., This work was supported by MICIU/AEI/10.13039/50110001 1033 and ERDF/EU under Grant PID2021-122136OB-C21, and by the Federal Ministry of Education and Research of Germany and the state of North-Rhine Westphalia as part of the Lamarr Institute for Machine Learning and Artificial Intelligence (Lamarr22B). Mohammadreza Javadiha was also supported by the Spanish Ministry of Science and Innovation and ERDF/EU under Grant PRE2018-086835. The ViRVIG group also received funding from the Department of Research and Universities of the Government of Catalonia (2021 SGR 01035)., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Evaluating LLMs’ abilities to create charts, a systematic approach

  • Ribalta Albado, Maria
  • Vázquez Alcocer, Pere Pau
The use of generative models, especially those based on pretrained transformers, has become a common practice in code development. Tools such as GitHub Copilot, Cursor, and the direct use of conversational chatbots have proven useful to accelerate the development of applications. Unfortunately, generative models are unable to determine what is correct or wrong, and their outputs may contain errors. Their stochastic nature does not guarantee a single solution for the same problem, either. Furthermore, the output depends largely on the prompt issued by the user. To assess the capabilities of LLMs, some benchmarks have been proposed. Unfortunately, they often rely on ground truth data that may not be available. As a result, the extent to which modern LLMs can create charts needs further investigation. This work contributes to the understanding of the generative models’ ability to create charts in three ways: (a) Creating a dataset of prompts, data sources, and chart types to analyze, (b) Designing a set of systematic experiments that cover a wide range of commonly used charts, and variations of the visual variables, and (c) by empirically analyzing the performance of a large set of LLMs of different sizes, including Claude, CodeLlama, Gemini, Gemma, GPT4o, Llama 3.1, and Mixtral. Our results indicate that even the most advanced LLMs have room for improvement., This project has been supported by PID2021-122136OB-C21 from the Ministerio de Ciencia e Innovación, Spain, by 839 FEDER (EU) funds, and 2021 SGR 01035 by Generalitat de Catalunya, Spain., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Revealing the layers of history: enriching medieval cultural heritage 4D models with interactive narratives

  • Andújar Gran, Carlos Antonio
  • Cayuela Vellido, Begoña
  • Lorès Otzet, Immaculada
  • Bosch Geli, Carles
  • Guardia Pons, Milagros
  • Muñoz Pandiella, Imanol
Cultural heritage (CH) preservation is increasingly leveraging digital technologies to support art history research and provide educational experiences to the public. In the case of some Medieval monuments, the myriad of architectural and pictorial transformations suffered due to the passage of time, and the geographical dispersion of its remaining decoration substantially hinders their understanding. In this article, we present a novel approach for enriching multi-layered, multi-phase 3D models of CH monuments with art history knowledge. Such knowledge is presented in the form of image-based annotations, interactive narratives, points of interest, text, and audio, organized into a collection of guided tours enabling visitors to traverse through different historical phases of the monument. A key benefit of our approach is that art historians require no programming skills to easily define all the elements of the interactive narratives. Our approach integrates various multimedia elements seamlessly into 3D models, offering visitors an engaging and informative journey through space and time. We conducted a pilot user evaluation to demonstrate the effectiveness of our approach, highlighting its value in enhancing visitor engagement and understanding of CH. The study also provided valuable feedback on the user experience, which allowed us to enhance the final version of the web-based application and the associated narratives., This work has been partially supported by projects PID2021-122136OB-C21 and PID2021-122136OB-C22 funded by MCIN/AEI/10.13039/501100011033 and ERDF “A way of making Europe,” by the EU Horizon 2020, JPICH Conservation, Protection and Use initiative (JPICH-0127), the Spanish Agencia Estatal de Investigación (grant PCI2020-111979), and the Department of Research and Universities of the Government of Catalonia (2021 SGR 01035)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Triangle influence supersets for fast distance computation

  • Pujol Puig, Eduard
  • Chica Calaf, Antonio
We present an acceleration structure to efficiently query the Signed Distance Field (SDF) of volumes represented by trianglemeshes. The method is based on a discretization of space. In each node, we store the triangles defining the SDF behaviour inthat region. Consequently, we reduce the cost of the nearest triangle search, prioritizing query performance, while avoidingapproximations of the field. We propose a method to conservatively compute the set of triangles influencing each node. Given anode, each triangle defines a region of space such that all points inside it are closer to a point in the node than the triangle is.This property is used to build the SDF acceleration structure. We do not need to explicitly compute these regions, which is crucialto the performance of our approach. We prove the correctness of the proposed method and compare it to similar approaches,confirming that our method produces faster query times than other exact methods., This work has been partially funded by Ministeri de Ciència i Innovació (MICIN), Agencia Estatal de Investigación (AEI) and the Fons Europeu de Desenvolupament Regional (FEDER) (project PID2021-122136OB-C21 funded by MCIN/AEI/10.13039/501100011033/FEDER, UE). The first author gratefully acknowledges the Universitat Politècnica de Catalunya and Banco Santander for the financial support of his predoctoral grant FPI-UPC grant., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Fitted avatars: automatic skeleton adjustment for self-avatars in virtual reality

  • Pontón Martínez, José Luis
  • Ceballos Inza, Víctor
  • Acosta Argueta, Lesly María
  • Ríos Jerez, Alejandro
  • Monclús Lahoya, Eva
  • Pelechano Gómez, Núria
In the era of the metaverse, self-avatars are gaining popularity, as they can enhance presence and provide embodiment when a user is immersed in Virtual Reality. They are also very important in collaborative Virtual Reality to improve communication through gestures. Whether we are using a complex motion capture solution or a few trackers with inverse kinematics (IK), it is essential to have a good match in size between the avatar and the user, as otherwise mismatches in self-avatar posture could be noticeable for the user. To achieve such a correct match in dimensions, a manual process is often required, with the need for a second person to take measurements of body limbs and introduce them into the system. This process can be time-consuming, and prone to errors. In this paper, we propose an automatic measuring method that simply requires the user to do a small set of exercises while wearing a Head-Mounted Display (HMD), two hand controllers, and three trackers. Our work provides an affordable and quick method to automatically extract user measurements and adjust the virtual humanoid skeleton to the exact dimensions. Our results show that our method can reduce the misalignment produced by the IK system when compared to other solutions that simply apply a uniform scaling to an avatar based on the height of the HMD, and make assumptions about the locations of joints with respect to the trackers., This work was funded by the Spanish Ministry of Science and Innovation (PID2021-122136OB-C21). Jose Luis Ponton was also funded by the Spanish Ministry of Universities (FPU21/01927)., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

A highly-configurable session designer for VR nursing training

  • Andújar Gran, Carlos Antonio
  • Chica Calaf, Antonio
  • Fairén González, Marta
  • Garcia Toledo, Òscar
  • Nieto Maldonado, Jordi
  • Tortosa Ortiz-Villajos, Sergi
  • Insa Calderón, Esther
Virtual Reality (VR) has proven to be a valuable tool for medical and nursing education. VR simulators are available at any time and from anywhere, and can be used with or without faculty supervision, which results in a significant optimization of time, space, and resources. In this paper we present a highly-configurable session designer for VR-based nursing education following the Standards for QUality Improvement Reporting Excellence: SQUIRE 2.0 and SQUIRE-EDU. Unlike existing platforms, we focus on letting educators quickly customize the training sessions in multiple aspects. This is achieved through the use of a visual editor, where educators can choose the environment and the elements in the session, and a node-based visual system for programming the behavior of the sessions. Thanks to end-user customization, educators can develop different variants of a session, adapt the sessions to new nursing protocols, incorporate new equipment or instruments, or accommodate custom environments. We have implemented the designer and tested it on various nursing training scenarios. The designer can be integrated into VR simulators to help educators save time in delivering highly customizable training., This research was partially supported by MICIU / AEI / 10.13039/501100011033 and ERDF / EU under Grant PID2021- 122136OB-C21., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Exploring the role of expected collision feedback in crowded virtual environments

  • Yun, Haoran
  • Pontón Martínez, José Luis
  • Beacco Porres, Alejandro
  • Andújar Gran, Carlos Antonio
  • Pelechano Gómez, Núria
An increasing number of virtual reality applications require environments that emulate real-world conditions. These environments often involve dynamic virtual humans showing realistic behaviors. Understanding user perception and navigation among these virtual agents is key for designing realistic and effective environments featuring groups of virtual humans. While collision risk significantly influences human locomotion in the real world, this risk is largely absent in virtual settings. This paper studies the impact of the expected collision feedback on user perception and interaction with virtual crowds. We examine the effectiveness of commonly used collision feedback techniques (auditory cues and tactile vibrations) as well as inducing participants to expect that a physical bump with a real person might occur, as if some virtual humans actually correspond to real persons embodied into them and sharing the same physical space. Our results indicate that the expected collision feedback significantly influences both participant behavior—encompassing global navigation and local movements—and subjective perceptions of presence and copresence. Specifically, the introduction of a perceived risk of actual collision was found to significantly impact global navigation strategies and increase the sense of presence. Auditory cues had a similar effect on global navigation and additionally enhanced the sense of copresence. In contrast, vibrotactile feedback was primarily effective in influencing local movements., This work has received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 860768 (CLIPE project), and from MCIN/AEI/10.13039/501100011033/FEDER, UE (PID2021- 122136OB-C21), ”A way to make Europe”. Jose Luis Ponton was also funded by the Spanish Ministry of Universities (FPU21/01927)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Two-step techniques for accurate selection of small elements in VR environments

  • Molina López, Elena
  • Vázquez Alcocer, Pere Pau
One of the key interactions in 3D environments is target acquisition, which can be challenging when targets are small or in cluttered scenes. Here, incorrect elements may be selected, leading to frustration and wasted time. The accuracy is further hindered by the physical act of selection itself, typically involving pressing a button. This action reduces stability, increasing the likelihood of erroneous target acquisition. We focused on molecular visualization and on the challenge of selecting atoms, rendered as small spheres. We present two techniques that improve upon previous progressive selection techniques. They facilitate the acquisition of neighbors after an initial selection, providing a more comfortable experience compared to using classical ray-based selection, particularly with occluded elements. We conducted a pilot study followed by two formal user studies. The results indicated that our approaches were highly appreciated by the participants. These techniques could be suitable for other crowded environments as well., This paper has been supported by TIN2017-88515-C2-1-R (GEN3DLIVE), from the Spanish Ministerio de Economía y Competitividad and PID2021-122136OB-C21 from the Ministerio de Ciencia e Innovación, Spain, by 839 FEDER (EU) funds. Elena Molina has been supported by FI-SDUR doctoral grant from Generalitat de Catalunya, and FPU grant from the Ministerio de Ciencia e Innovación, Spain ., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

DRL-driven padel players: Simulating padel matches through deep reinforcement learning in real and hypothetical scenarios

  • Javadiha, Mohammadreza
  • Ji, Jia Long
  • Zhou, Wenqi
  • Lacasa Claver, Enrique
  • Andújar Gran, Carlos Antonio
Recent advances in Deep Reinforcement Learning (DRL) have opened new avenues for sport research. DRL allows virtual agents to learn and solve complex tasks with minimal input, which means that models can be trained with little or no data collection. This enables the creation of sport simulations that go beyond the replication of real matches, exploring entirely new conditions where prior knowledge is limited. In this paper, we investigate the generation of simulated data for padel matches, aiming to either replicate real environments or investigate hypothetical ones with altered parameters such as non-standard court sizes. We focus on high-level behavioural aspects of the players rather than low-level skills. We present a proof-of-concept implementation of a DRL system that enables robot-like agents to learn and play padel under varying conditions. After training different types of virtual agents, we observed their overall behaviour during simulated matches. In particular, for certain combinations of parameters, the agents learned to coordinate their movements similarly to professional players. We argue that DRL-driven players have the potential to complement traditional mathematical models, especially in scenarios where data collection is challenging or unfeasible. This could lead to more engaging, safer and more inclusive sports in the future., This work was supported by the MICIU/AEI/10.13039/501100011033 and ERDF/EU under [Grant PID2021-122136OB-C21], and the Department of Research andUniversities of the Government of Catalonia [2021 SGR 01035]., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

PADELVIC: Multicamera videos and motion capture data in padel matches, PADELVIC: Videos multicámara y datos de captura de movimiento de un partido de pádel amateur

  • Javadiha, Mohammadreza
  • Andújar Gran, Carlos Antonio
  • Calvanese, Michele
  • Lacasa Claver, Enrique
  • Moyes Ardiaca, Jordi
  • Pontón Martínez, José Luis
  • Susín Sánchez, Antonio
  • Wang, Jiabo
Recent advances in computer vision and deep learning techniques have opened new possibilities regarding the automatic labeling of sport videos. However, an essen-tial requirement for supervised techniques is the availability of accurately labeled training datasets. In this paper we present PadelVic, an annotated dataset of an ama-teur padel match which consists of multi-view video streams, estimated positional data for all four players within the court (and for one of the players, accurate motion capture data of his body pose), as well as synthetic videos specifically designed to serve as training sets for neural networks estimating positional data from videos. For the recorded data, player positions were estimated by applying a state-of-the-art pose estimation technique to one of the videos, which yields a relatively small positional error (M=16 cm, SD=13 cm). For one of the players, we used a motion capture system providing the orientation of the body parts with an accuracy of 1.5º RMS. The highest accuracy though comes from our synthetic dataset, which provides ground-truth po-sitional and pose data of virtual players animated with the motion capture data. As an example application of the synthetic dataset, we present a system for a more accurate prediction of the center-of-mass of the players projected onto the court plane, from a single-view video of the match. We also discuss how to exploit per-frame positional data of the players for tasks such as synergy analysis, collective tactical analysis, and player profile generation., This work has received funding from project PID2021-122136OB-C21 funded by MCIN/AEI and FEDER “A way to make Europe”. Mohammadreza Javadiha and Jose Luis Ponton were also funded by the Spanish Ministry of Science, Innovation and Universities, grants PRE2018-086835 and FPU21/01927., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Step2Motion: Locomotion reconstruction from pressure sensing insoles

  • Pontón Martínez, José Luis
  • Alvarado Pinero, Eduardo
  • Foo, LinGeng
  • Pelechano Gómez, Núria
  • Andújar Gran, Carlos Antonio
  • Habermann, Marc
Human motion is fundamentally driven by continuous physical interaction with the environment. Whether walking, running, or simply standing, the forces exchanged between our feet and the ground provide crucial insights for understanding and reconstructing human movement. Recent advances in wearable insole devices offer a compelling solution for capturing these forces in diverse, real-world scenarios. Sensor insoles pose no constraint on the users' motion (unlike mocap suits) and are unaffected by line-of-sight limitations (in contrast to optical systems). These qualities make sensor insoles an ideal choice for robust, unconstrained motion capture, particularly in outdoor environments. Surprisingly, leveraging these devices with recent motion reconstruction methods remains largely unexplored. Aiming to fill this gap, we present Step2Motion, the first approach to reconstruct human locomotion from multi-modal insole sensors. Our method utilizes pressure and inertial data—accelerations and angular rates—captured by the insoles to reconstruct human motion. We evaluate the effectiveness of our approach across a range of experiments to show its versatility for diverse locomotion styles, from simple ones like walking or jogging up to moving sideways, on tiptoes, slightly crouching, or dancing. The complete source code, trained model, data, and supplementary material used in this paper can be found at: https://vcai.mpi-inf.mpg.de/projects/Step2Motion/, This work has received funding from MCIN/AEI/10.13039/501100011033/FEDER, UE (Spain) in the framework of the project PID2021-122136OB-C21, and with the support of the Department of Research and Universities of the Government of Catalonia (2021 SGR 01035). Jose Luis Ponton was also funded by the Spanish Ministry of Universities (FPU21/01927 andEST24/00555)., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Improving dimensionality reduction projections for data visualization

  • Rafieian, Bardia
  • Hermosilla Casajús, Pedro
  • Vázquez Alcocer, Pere Pau
In data science and visualization, dimensionality reduction techniques have been extensively employed for exploring large datasets. These techniques involve the transformation of high-dimensional data into reduced versions, typically in 2D, with the aim of preserving significant properties from the original data. Many dimensionality reduction algorithms exist, and nonlinear approaches such as the t-SNE (t-Distributed Stochastic Neighbor Embedding) and UMAP (Uniform Manifold Approximation and Projection) have gained popularity in the field of information visualization. In this paper, we introduce a simple yet powerful manipulation for vector datasets that modifies their values based on weight frequencies. This technique significantly improves the results of the dimensionality reduction algorithms across various scenarios. To demonstrate the efficacy of our methodology, we conduct an analysis on a collection of well-known labeled datasets. The results demonstrate improved clustering performance when attempting to classify the data in the reduced space. Our proposal presents a comprehensive and adaptable approach to enhance the outcomes of dimensionality reduction for visual data exploration., This research was funded by PID2021-122136OB-C21 from the Ministerio de Ciencia e Innovación, Spain, by 839 FEDER (EU) funds., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Monkey see, monkey break? Study of rule-breaking imitation in virtual crowds

  • Huang, Kexiang
  • Yin, Tairan
  • Pontón Martínez, José Luis
  • Tang, Ruida
  • Itatani, Reiya
  • Ding, Gangyi
  • Argudo Medrano, Óscar
  • Pelechano Gómez, Núria
Rule-breaking behaviors, such as jaywalking or skipping queues, are common in crowds but difficult to study in real-world settings due to limited control and observability. Virtual reality (VR) provides a controlled alternative, but its validity depends on whether VRelicits realistic rule-breaking behavior. We conducted a VR study with 65 participants navigating a virtual city with four scenarios differing in social norms: walking on grass, crossing outside a crosswalk, jaywalking at a red light, and skipping a line. In each scenario, the proportion of rule-breaking virtual characters was manipulated (0%, 25%, 50%). Participants' movements and gaze were recorded to assess behavior and attention. Results showed higher rule-breaking as the number of violators increased, except in the low-stakes crossing scenario. Rule-breakers attended more to violating characters, whereas rule-followers focused on compliant ones. Participants cited efficiency, safety, and social norms as key factors guiding their decisions. Overall, VR reproduced natural patterns of social compliance and noncompliance, supporting its use for studying rule-breaking and applications in crowd simulation, urban design, safety training, and immersive media., This work was supported by grants PID2021-122136OB-C21 from MCIN/AEI/10.13039/501100011033/FEDER “A way to make Europe” and the National Natural Science Foundation of China (No.62177005), Peer Reviewed, 11 - Ciutats i Comunitats Sostenibles, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Accurate molecular atom selection in VR

  • Molina López, Elena
  • Vázquez Alcocer, Pere Pau
Target acquisition is a basic task that is part of almost any high-level interaction in 3D environments. Therefore, providing accurate selection is a necessity for most applications and games. When targets are small and scenes are cluttered, selection becomes inaccurate. This may lead to selecting the wrong elements which, apart from the time consumed, may become a frustrating experience. Besides the unintentional tremor, the button/trigger press for effectively selecting an element further reduces our stability, increasing the probability of an incorrect target acquisition. In this paper, we focus on molecular visualization and address the problem of selecting atoms, which are rendered as small spheres. We build upon previous progressive selection algorithms and present two alternatives that accelerate the selection of neighbors after an initial selection. We have implemented and analyzed such techniques through a formal user study and found that they were highly appreciated by the users. These selection methods may be suitable for other crowded environments beyond molecular visualization., This paper has been supported by TIN2017-88515-C2-1- R (GEN3DLIVE), from the Spanish Ministerio de Economía y Competitividad and PID2021-122136OB-C21 from the Ministerio de Ciencia e Innovación, by 839 FEDER (EU) funds. Elena Molina has been supported by FI-SDUR doctoral grant from Generalitat de Catalunya, and FPU grant from the Ministerio de Ciencia e Innovación., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Enriching medieval 4D models with interactive narratives

  • Andújar Gran, Carlos Antonio
  • Cayuela Vellido, Begoña
  • Lorès Otzet, Immaculada
  • Bosch Geli, Carles
  • Guardia Pons, Milagros
  • Muñoz Pandiella, Imanol
Digital technologies are increasingly used in cultural heritage preservation and dissemination, but some monuments pose significant challenges due to substantial historical architectural and pictorial changes. This paper introduces a method to enrich multi-layered, multi-phase 3D models of cultural heritage with art history knowledge. This is achieved through imagebased annotations, interactive narratives, points of interest, text, and audio, creating guided tours of the monument's historical phases. The approach is user-friendly for art historians, requiring no programming skills for defining such elements. A pilot user evaluation demonstrated its effectiveness in enhancing visitor engagement and understanding, and allowed improving the final web-based application and narratives., This work has been partially supported by projects PID2021-122136OB-C21 and PID2021-122136OB-C22 funded by MCIN/AEI/ 10.13039/501100011033 and ERDF "A way of making Europe", by the EU Horizon 2020, JPICH Conservation, Protection and Use initiative (JPICH-0127), the Spanish Agencia Estatal de Investigación (grant PCI2020-111979) and by the Department of Research and Universities of the Government of Catalonia (2021 SGR 01035)., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Padelvic: Multicamera videos and motion capture data in padel matches, PADELVIC: Videos multicámara y datos de captura de movimiento de un partido de pádel amateur

  • Javadiha, Mohammadreza
  • Andujar, Carlos
  • Calvanese, Michele
  • Lacasa Claver, Enrique
  • Moyés, Jordi
  • Pontón, José Luis
  • Susin, Antonio
  • Wang, Jiabo
Los recientes avances en visión por computador y las técnicas de aprendizaje profundo han abierto nuevas posibilidades en cuanto a la etiquetación automática de videos deportivos. Sin embargo, un requisito esencial para las técnicas de aprendizaje supervisado es la disponibilidad de extensos conjuntos de entrenamiento con anotaciones precisas. En este artículo presentamos PadelVic, un conjunto de datos anotado de un partido amateur de pádel que incluye múltiples secuencias de video desde diferentes ángulos, datos precisos de captura de movimiento de uno de los jugadores, así como videos sintéticos diseñados específicamente para ser utilizados como conjuntos de entrenamiento para redes neuronales convolucionales que estiman datos de posición a partir de videos. Como demostración de una de las aplicaciones del conjunto de datos, presentamos un sistema para predecir con precisión el centro de masa de los jugadores proyectado sobre el plano de la pista, a partir de un solo video del partido., Recent advances in computer vision and deep learning techniques have opened new possibilities regarding the automatic labeling of sport videos. However, an essential requirement for supervised techniques is the availability of accurately labeled training datasets. In this paper we present PadelVic, an annotated dataset of an amateur padel match which consists of multi-view video streams, estimated positional data for all four players within the court (and for one of the players, accurate motion capture data of his body pose), as well as synthetic videos specifically designed to serve as training sets for neural networks estimating positional data from videos. For the recorded data, player positions were estimated by applying a state-of-the-art pose estimation technique to one of the videos, which yields a relatively small positional error (M=16 cm, SD=13 cm). For one of the players, we used a motion capture system providing the orientation of the body parts with an accuracy of 1.5º RMS. The highest accuracy though comes from our synthetic dataset, which provides ground-truth positional and pose data of virtual players animated with the motion capture data. As an example application of the synthetic dataset, we present a system for a more accurate prediction of the center-of-mass of the players projected onto the court plane, from a single-view video of the match. We also discuss how to exploit per-frame positional data of the players for tasks such as synergy analysis, collective tactical analysis, and player profile generation., This work has received funding from project PID2021-122136OB-C21 funded by MCIN/AEI and FEDER “A way to make Europe”. Mohammadreza Javadiha and Jose Luis Ponton were also funded by the Spanish Ministry of Science, Innovation and Universities, grants PRE2018-086835 and FPU21/01927
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

GREIL-Crowds: crowd simulation with deep reinforcement learning and examples

  • Charalambous, Panayiotis
  • Pettré, Julien
  • Vassiliades, Vassilis
  • Chrysanthou, Yiorgos
  • Pelechano Gómez, Núria
Simulating crowds with realistic behaviors is a difficult but very important task for a variety of applications. Quantifying how a person balances between different conflicting criteria such as goal seeking, collision avoidance and moving within a group is not intuitive, especially if we consider that behaviors differ largely between people. Inspired by recent advances in Deep Reinforcement Learning, we propose Guided REinforcement Learning (GREIL) Crowds, a method that learns a model for pedestrian behaviors which is guided by reference crowd data. The model successfully captures behaviors such as goal seeking, being part of consistent groups without the need to define explicit relationships and wandering around seemingly without a specific purpose. Two fundamental concepts are important in achieving these results: (a) the per agent state representation and (b) the reward function. The agent state is a temporal representation of the situation around each agent. The reward function is based on the idea that people try to move in situations/states in which they feel comfortable in. Therefore, in order for agents to stay in a comfortable state space, we first obtain a distribution of states extracted from real crowd data; then we evaluate states based on how much of an outlier they are compared to such a distribution. We demonstrate that our system can capture and simulate many complex and subtle crowd interactions in varied scenarios. Additionally, the proposed method generalizes to unseen situations, generates consistent behaviors and does not suffer from the limitations of other data-driven and reinforcement learning approaches., This project has received funding from the European Union’s Horizon 2020 Research and Innovation Programme under Grant Agreement No 739578 and the Government of the Republic of Cyprus through the Deputy Ministry of Research, Innovation and Digital Policy. This work has received funding from the Spanish Ministry of Science and Innovation (PID2021-122136OB-C21)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

A 3D feature-based approach for mapping scaling effects on stone monuments

  • Muñoz Pandiella, Imanol
  • Pueyo Sandez, Francesc Xavier
  • Bosch Geli, Carles
Weathering effects caused by physical, chemical, or biological processes result in visible damages that alter the appearance of stones’ surfaces. Consequently, weathered stone monuments can offer a distorted perception of the artworks to the point of making their interpretation misleading. Being able to detect and monitor decay is crucial for restorers and curators to perform important tasks such as identifying missing parts, assessing the preservation state, or evaluating curating strategies. Decay mapping, the process of identifying weathered zones of artworks, is essential for preservation and research projects. This is usually carried out by marking the affected parts of the monument on a 2D drawing or picture of it. One of the main problems of this methodology is that it is manual work based only on experts’ observations. This makes the process slow and often results in disparities between the mappings of the same monument made by different experts. In this paper, we focus on the weathering effect known as “scaling”, following the ICOMOS ISCS definition. We present a novel technique for detecting, segmenting, and classifying these effects on stone monuments. Our method is user-friendly, requiring minimal user input. By analyzing 3D reconstructed data considering geometry and appearance, the method identifies scaling features and segments weathered regions, classifying them by scaling subtype. It shows improvements over previous approaches and is well-received by experts, representing a significant step towards objective stone decay mapping., This work has been partially supported by projects PID2021-122136OB-C21 and PID2021-122136OB-C22 funded by MCIN/AEI/10.13039/501100011033 and by ERDF A way of making Europe and developed in the context of Grup de Recerca Consolidat ViRVIG 2021-SGR-01035 with the support of the Departament of Research and Universities of the Generalitat de Catalunya., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Revealing the Layers of History: Enriching Medieval Cultural Heritage 4D Models with Interactive Narratives

  • Andujar, Carlos
  • Cayuela Vellido, Begoña
  • Lorés, Immaculada
  • Bosch Geli, Carles
  • Guardia Pons, Milagros
  • Muñoz-Pandiella, Imanol
Cultural heritage (CH) preservation is increasingly leveraging digital technologies to support art history research and provide educational experiences to the public. In the case of some Medieval monuments, the myriad of architectural and pictorial transformations suffered due to the passage of time, and the geographical dispersion of its remaining decoration substantially hinders their understanding. In this article, we present a novel approach for enriching multi-layered, multi-phase 3D models of CH monuments with art history knowledge. Such knowledge is presented in the form of image-based annotations, interactive narratives, points of interest, text, and audio, organized into a collection of guided tours enabling visitors to traverse through different historical phases of the monument. A key benefit of our approach is that art historians require no programming skills to easily define all the elements of the interactive narratives. Our approach integrates various multimedia elements seamlessly into 3D models, offering visitors an engaging and informative journey through space and time. We conducted a pilot user evaluation to demonstrate the effectiveness of our approach, highlighting its value in enhancing visitor engagement and understanding of CH. The study also provided valuable feedback on the user experience, which allowed us to enhance the final version of the web-based application and the associated narratives., This work has been partially supported by projects PID2021-122136OB-C21 and PID2021-122136OB-C22 funded by MCIN/ AEI/10.13039/501100011033 and ERDF “A way of making Europe,” by the EU Horizon 2020, JPICH Conservation, Protection and Use initiative (JPICH-0127), the Spanish Agencia Estatal de Investigación (grant PCI2020-111979), and the Department of Research and Universities of the Government of Catalonia (2021 SGR 01035).
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

PHA: Part-wise heterogeneous agents with reusable policy priors for physics-based motion synthesis

  • Carranza Cobeñas, Luis Alberto
  • Argudo Medrano, Óscar
  • Andújar Gran, Carlos Antonio
Performing everyday tasks requires both large-scale body movements driven by the arms, legs, and torso, and fine motor skills, particularly in the hands. However, existing reinforcement learning approaches often struggle to efficiently acquire these diverse motion skills from scratch, leading to slow convergence or suboptimal policies. We observed that the motor skills of distinct body parts exhibit a certain level of independence. This suggests the potential advantage of independently pretraining specific dexterous limbs prior to their integration in full-body motion tasks. Inspired by this, we present Part-wise Heterogeneous Agents (PHA), a cooperative multi-agent reinforcement learning approach where body parts are treated as independent agents, allowing for specialized skill acquisition and cooperative execution of complex full-body tasks. Furthermore, our method enables the pretraining of fine motor skills, such as gripping a bar or grabbing a climbing hold, before integrating them with other body parts for complex whole-body coordination, thus introducing part-wise Reusable Policy Priors. We tested our technique on challenging tasks such as rope climbing, rock bouldering and traversing a horizontal ladder. Our approach not only accelerates convergence, but also improves overall policy quality, achieving motion tasks that single-agent approaches struggle to solve. Our results also demonstrate adaptability, enabling Reusable Policy Priors to adjust their policies to successfully perform complex tasks in scenarios not seen during training., This work was supported by Grant PID2021-122136OB-C21, funded by MICIU/AEI/10.13039/501100011033 and ERDF/EU, as well as the Department of Research and Universities of the Government of Catalonia (2021 SGR 01035). Luis Carranza was also supported with the grant PRE2022-102912 funded by MICIU/AEI/10.13039/501100011033 and ESF+., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Improved multi-label hierarchical patent classification using LLMs

  • Rafieian, Bardia
  • Vázquez Alcocer, Pere Pau
Classifying multi-label documents has always been a challenging task, especially when the labels follow a hierarchical structure. This complexity increases the difficulty of accurately predicting multiple interrelated labels, often limiting the overall classification performance. To address these challenges and improve the accuracy of hierarchical multi-label classification systems, we introduce a novel pipeline leveraging large language models (LLMs). By incorporating quantization techniques and optimizing weight updates through smaller matrices, we improve computational efficiency and scalability. Our approach demonstrates an improvement in accuracy compared to current state-of-the-art models. In this work, we apply this pipeline to patent documents, focusing on the multi-label hierarchical text classification problem using a transformer-based architecture. The hierarchical structure of the CPC (Cooperative Patent Classification) labels is preserved through a graph-based taxonomy, which enables more effective processing of patent categories. Our model is trained and evaluated on the USPTO-70k dataset, and we achieve substantial improvements across various metrics, including precision, recall, F1-score, and AUC., This project has been supported by PID2021-122136OB-C21 from the Ministerio de Ciencia e Innovación, by 839 FEDER (EU) funds., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Digital reconstruction of partially lost altarpieces. The case of the Rosary's altarpiece of Sant Pere Màrtir de Manresa

  • Pagès Vilà, Anna
  • Pueyo Sandez, Francesc Xavier
  • Muñoz Pandiella, Imanol
© 2023 Author's | Association for Computing Machinery. This is the author's version of the work. It is posted here for your personal use. Not for redistribution. The definitive Version of Record was published in Journal on Computing and Cultural Heritage, http://dx.doi.org/10.1145/3652860., An important challenge of Digital Cultural Heritage is to contribute to the recovery of artworks with their original shape and appearance. Many altarpieces, which are very relevant Christian art elements, have been damaged and/or partly or fully lost. Therefore, the only way to recover them is to carry out their digital reconstruction. Although the procedure that we present here is valid for any altarpiece with similar characteristics, and even for other akin elements, our test bench is the altarpieces that were damaged or destroyed or that disappeared during the Spanish Civil War (1936–1939) in Catalonia, where most suffered these effects. The first step of our work has been the classification of these artworks into different categories on the basis of their degree of destruction and of the available visual information related to each one.</jats:p> <jats:p>This article proposes, for the first time to our knowledge, a workflow for the virtual reconstruction, through photogrammetry, digital modeling, and digital color restoration, of whole altarpieces partially preserved with very little visual information. Our case study is the Rosary’s altarpiece of Sant Pere Màrtir de Manresa church. Currently, this altarpiece is partially preserved in fragments in the Museu Comarcal de Manresa (Spain). But it cannot be reassembled physically owing to the lack of space (actually the church does not exist anymore) and the cost of such an operation. Thus, there is no other solution than the digital one to contemplate and study the altarpiece as a whole. The reconstruction that we provide allows art historians and the general public to virtually see the altarpiece complete and assembled as it was until 1936. The results obtained also allow us to see in detail the reliefs and ornaments of the altarpiece with their digitally restored color., The authors would like to thank Dr. Joan Bosch for his support and advice during the development of the research presented here and the Museu de Manresa - Museu del Barroc de Catalunya for granting access to the artworks of the Rosary’s altarpiece and permitting their digitization, and authorizing us their use. We also want to express our gratitude to the GIRAVOLT program (https://patrimoni.gencat.cat/en/giravolt) of the Departament de Cultura de la Generalitat de Catalunya for providing us with the photogrammetric models of the scenes of the Annunciation and Pentecost of the altarpiece and granting us their use. This work is part of the doctoral thesis of the first author, which is funded by the IFUdG2022-2024 grant, of the University of Girona with the collaboration of the Santander Bank. It has been partially supported by the projects PID2021-122136OB-C21 and PID2021-122136OB-C22 funded by MCIN/AEI/ 10.13039/501100011033 and by ERDF A way of making Europe and by the project PID2020-114339GB-C21 funded by the MCIN/AEI/, and developed in the context of Grup de Recerca Consolidat ViRVIG 2021-SGR-01035 with the support of the Departament of Research and Universities of the Generalitat de Catalunya., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Bayesian 3D shape reconstruction from noisy points and normals

  • Pujol Puig, Eduard
  • Chica Calaf, Antonio
Reconstructing three-dimensional shapes from point clouds remains a central challenge in geometry processing, particularly due to the inherent uncertainties in real-world data acquisition. In this work, we introduce a novel Bayesian framework that explicitly models and propagates uncertainty from both input points and their estimated normals. Our method incorporates the uncertainty of normals derived via Principal Component Analysis (PCA) from noisy input points. Building upon the Smooth Signed Distance (SSD) reconstruction algorithm, we integrate a smoothness prior based on the curvatures of the resulting implicit function following Gaussian behavior. Our method reconstructs a shape represented as a distribution, from which sampling and statistical queries regarding the shape's properties are possible. Additionally, because of the high cost of computing the variance of the resulting distribution, we develop efficient techniques for variance computation. Our approach thus combines two common steps of the geometry processing pipeline, normal estimation and surface reconstruction, while computing the uncertainty of the output of each of these steps., This work has been partially funded by Ministeri de Ciència i Innovació (MICIN), Agencia Estatal de Investigación(AEI) and the Fons Europeu de Desenvolupament Regional (FEDER) (project PID2021-122136OB-C21 funded by MCIN/AEI/10.13039/501100011033/FEDER, UE); and by the Department of Research and Universities of the Government of Catalonia (2021 SGR 01035). The first author gratefully acknowledges the Universitat Politècnica de Catalunya and Banco Santander for the financial support of his predoctoral grant FPI-UPC grant. Open Access funding enabled and organized by Universitat Politècnica de Catalunya., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

A highly-configurable session designer for VR nursing training

  • Andújar, Carlos
  • Chica, Antoni
  • Fairén, Marta
  • García, Óscar
  • Nieto, Jordi
  • Tortosa, Sergi
  • Insa-Calderón, Esther
Virtual Reality (VR) has proven to be a valuable tool for medical and nursing education. VR simulators are available at any time and from anywhere, and can be used with or without faculty supervision, which results in a significant optimization of time, space, and resources. In this paper we present a highly-configurable session designer for VR-based nursing education following the Standards for QUality Improvement Reporting Excellence: SQUIRE 2.0 and SQUIRE-EDU. Unlike existing platforms, we focus on letting educators quickly customize the training sessions in multiple aspects. This is achieved through the use of a visual editor, where educators can choose the environment and the elements in the session, and a node-based visual system for programming the behavior of the sessions. Thanks to end-user customization, educators can develop different variants of a session, adapt the sessions to new nursing protocols, incorporate new equipment or instruments, or accommodate custom environments. We have implemented the designer and tested it on various nursing training scenarios. The designer can be integrated into VR simulators to help educators save time in delivering highly customizable training., This research was partially supported by MICIU / AEI / 10.13039/501100011033 and ERDF / EU under Grant PID2021-122136OB-C21.
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Reconstructing lost altarpieces: a differentiable rendering approach

  • Pagès Vilà, Anna
  • Muñoz Pandiella, Imanol
Studying works that have completely or partially disappeared is always difficult due to the lack of information. In more fortunate scenarios where photographs were taken before the destruction, the study of the piece is limited by the viewpoints captured in the available photographs. In this interdisciplinary research, we present a new methodology for reconstructing lost altarpieces from a single historical image, utilizing differentiable rendering techniques. We test our methodology by reconstructing some reliefs from the altarpiece of Sant Joan Baptista (Valls, Spain), which was destroyed in 1936. These results are valuable for both experts and the public, as they facilitate a better understanding of the relief's volumetrics and their spatial relationships, representing a significant advancement in the virtual recovery of lost artifacts., This work has been partially supported by the projects PID2021-122136OB-C21andPID2021-122136OB-C22fundedby MCIN/AEI/ 10.13039/501100011033 and by FEDER A way of making Europe, by the project PID2020-114339GB-C21 funded by the MCIN/AEI, and by the Department of Research and Universities of the Generalitat de Catalunya (2021 SGR 01035)., This work has been partially supported by the projects PID2021-122136OB-C21 and PID2021-122136OB-C22 funded by MCIN/AEI/ 10.13039/501100011033 and by FEDER A way of making Europe, by the project PID2020-114339GB-C21 funded by the MCIN/AEI, and by the Department of Research and Universities of the Generalitat de Catalunya (2021 SGR 01035)., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Features that influence bike sharing demand

  • Cortez Ordóñez, Alexandra Piedad
  • Vázquez Alcocer, Pere Pau
  • Sánchez Espigares, Josep Anton
During the last few years, Bike Sharing Systems (BSS) have become a popular means of transportation in several cities across the world, owing to their low costs and associated advantages. Citizens have adopted these systems as they help improve their health and contribute to creating more sustainable cities. However, customer satisfaction and the willingness to use the systems are directly affected by the ease of access to the docking stations and finding available bikes or slots. Therefore, system operators and managers' major responsibilities focus on urban and transport planning by improving the rebalancing operations of their BSS. Many approaches can be considered to overcome the unbalanced station problem, but predicting the number of arrivals and departures at the docking stations has been proven to be one of the most efficient. In this paper, we study the features that influence the prediction of bikes' arrivals and departures in Barcelona BSS, using a Random Forest model and a one-year data period. We considered features related to the weather, the stations' characteristics, and the facilities available within a 200-meter diameter of each station, called spatial features. The results indicate that features related to specific months, as well as temperature, pressure, altitude, and holidays, have a strong influence on the model, while spatial features have a small impact on the prediction results., This project has been supported by PID2021-122136OB-C21 from the Ministerio de Ciencia e Innovación, by 839 FEDER (EU) funds and 2021SGR00613-(ADBD) by Generalitat de Cataluña., Peer Reviewed, Objectius de Desenvolupament Sostenible::9 - Indústria, Innovació i Infraestructura, Objectius de Desenvolupament Sostenible::11 - Ciutats i Comunitats Sostenibles, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Animation fidelity in self-avatars: impact on user performance and sense of agency

  • Yun, Haoran
  • Pontón Martínez, José Luis
  • Andújar Gran, Carlos Antonio
  • Pelechano Gómez, Núria
The use of self-avatars is gaining popularity thanks to affordable VR headsets. Unfortunately, mainstream VR devices often use a small number of trackers and provide low-accuracy animations. Previous studies have shown that the Sense of Embodiment, and in particular the Sense of Agency, depends on the extent to which the avatar's movements mimic the user's movements. However, few works study such effect for tasks requiring a precise interaction with the environment, i.e., tasks that require accurate manipulation, precise foot stepping, or correct body poses. In these cases, users are likely to notice inconsistencies between their self-avatars and their actual pose. In this paper, we study the impact of the animation fidelity of the user avatar on a variety of tasks that focus on arm movement, leg movement and body posture. We compare three different animation techniques: two of them using Inverse Kinematics to reconstruct the pose from sparse input (6 trackers), and a third one using a professional motion capture system with 17 inertial sensors. We evaluate these animation techniques both quantitatively (completion time, unintentional collisions, pose accuracy) and qualitatively (Sense of Embodiment). Our results show that the animation quality affects the Sense of Embodiment. Inertial-based MoCap performs significantly better in mimicking body poses. Surprisingly, IK-based solutions using fewer sensors outperformed MoCap in tasks requiring accurate positioning, which we attribute to the higher latency and the positional drift that causes errors at the end-effectors, which are more noticeable in contact areas such as the feet., This project has received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No. 860768 (CLIPE project) and from MCIN/AEI/10.13039/501100011033/FEDER, UE (PID2021-122136OB-C21). Jose Luis Ponton was also funded by the Spanish Ministry of Universities (FPU21/01927)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

SparsePoser: Real-time full-body motion reconstruction from sparse data

  • Pontón Martínez, José Luis
  • Yun, Haoran
  • Aristidou, Andreas
  • Andújar Gran, Carlos Antonio
  • Pelechano Gómez, Núria
Accurate and reliable human motion reconstruction is crucial for creating natural interactions of full-body avatars in Virtual Reality (VR) and entertainment applications. As the Metaverse and social applications gain popularity, users are seeking cost-effective solutions to create full-body animations that are comparable in quality to those produced by commercial motion capture systems. In order to provide affordable solutions though, it is important to minimize the number of sensors attached to the subject’s body. Unfortunately, reconstructing the full-body pose from sparse data is a heavily under-determined problem. Some studies that use IMU sensors face challenges in reconstructing the pose due to positional drift and ambiguity of the poses. In recent years, some mainstream VR systems have released 6-degree-of-freedom (6-DoF) tracking devices providing positional and rotational information. Nevertheless, most solutions for reconstructing full-body poses rely on traditional inverse kinematics (IK) solutions, which often produce non-continuous and unnatural poses. In this article, we introduce SparsePoser, a novel deep learning-based solution for reconstructing a full-body pose from a reduced set of six tracking devices. Our system incorporates a convolutional-based autoencoder that synthesizes high-quality continuous human poses by learning the human motion manifold from motion capture data. Then, we employ a learned IK component, made of multiple lightweight feed-forward neural networks, to adjust the hands and feet toward the corresponding trackers. We extensively evaluate our method on publicly available motion capture datasets and with real-time live demos. We show that our method outperforms state-of-the-art techniques using IMU sensors or 6-DoF tracking devices, and can be used for users with different body dimensions and proportions., This work has received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie SkŁodowska-Curie grant agreement No 860768 (CLIPE project), HORIZON-CL4-2022-HUMAN-01 grant agreement No 101093159 (XR4ED), and from MCIN/AEI/10.13039/501100011033/FEDER, UE (PID2021-122136OB-C21). This project has also received funding from the European Union’s Horizon 2020 Research and Innovation Programme under Grant Agreement No 739578 and the Government of the Republic of Cyprus through the Deputy Ministry of Research, Innovation and Digital Policy. Jose Luis Ponton was also funded by the Spanish Ministry of Universities (FPU21/01927)., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

How to represent landmark trees in digital 3D maps? An automated workflow and user study

  • Grammatikaki, Angeliki
  • Eschner, Johannes
  • Ledermann, Florian
  • Argudo Medrano, Óscar
  • Waldner, Manuela
Digital 3D maps created from digital elevation models (DEMs) cannot properly capture trees due to the 2.5D nature of the DEMs. Leveraging publicly available DEMs and orthophotos as the only input data sources, we present a fully automatic pipeline that models landmark trees. We conducted two crowdsourced user studies to evaluate visual appeal and scene recognizability using two different levels of detail of tree representations generated using our pipeline. Users found highly detailed trees much more appealing, and also easier and more trustworthy to recognize, with open-ended responses revealing key factors like realism, coherence, and tree shape influencing their preferences. However, the ability to recognize a location seems to depend more on the surrounding environment than the representation of the landmark tree in focus. We discuss the implications of these results for digital 3D outdoor maps., The authors acknowledge TU Wien Bibliothek for financial support through its Open Access Funding Programme. O. Argudo also acknowledges Grant PID2021-122136OB-C21, funded by MICIU/AEI/10.13039/501100011033 and ERDF/EU, as well as the support of the Department of Research and Universities of the Government of Catalonia [2021 SGR 01035]., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Enriching Medieval 4D Models with Interactive Narratives

  • Andujar, Carlos
  • Cayuela Vellido, Begoña
  • Lorés, Immaculada
  • Bosch Geli, Carles
  • Guardia Pons, Milagros
  • Muñoz-Pandiella, Imanol
Digital technologies are increasingly used in cultural heritage preservation and dissemination, but some monuments pose significant challenges due to substantial historical architectural and pictorial changes. This paper introduces a method to enrich multi-layered, multi-phase 3D models of cultural heritage with art history knowledge. This is achieved through imagebased annotations, interactive narratives, points of interest, text, and audio, creating guided tours of the monument’s historical phases. The approach is user-friendly for art historians, requiring no programming skills for defining such elements. A pilot user evaluation demonstrated its effectiveness in enhancing visitor engagement and understanding, and allowed improving the final web-based application and narratives., This work has been partially supported by projects PID2021-122136OB-C21 and PID2021-122136OB-C22 funded by MCIN/AEI/ 10.13039/501100011033 and ERDF "A way of making Europe", by the EU Horizon 2020, JPICH Conservation, Protection and Use initiative (JPICH-0127), the Spanish Agencia Estatal de Investigación (grant PCI2020-111979) and by the Department of Research and Universities of the Government of Catalonia (2021 SGR 01035).
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Scalability evaluation of forecasting methods applied to bicycle sharing systems

  • Cortez Ordóñez, Alexandra Piedad
  • Vázquez Alcocer, Pere Pau
  • Sánchez Espigares, Josep Anton
Public Bicycle Sharing Systems have spread in many cities for the last decade. The need of analysis tools to predict the behavior or estimate balancing needs has fostered a wide set of approaches that consider many variables. Often, these approaches use a single scenario to evaluate their algorithms, and little is known about the applicability of such algorithms in cities of different sizes. In this paper, we evaluate the performance of widely known prediction algorithms for three sized scenarios: a small system, with around 20 docking stations, a medium-sized one, with 400+ docking stations, and a large one, with more than 1500 stations. The results show that Prophet and Random Forest are the prediction algorithms with more consistent results, and that small systems often have not enough data for the algorithms to perform a solid work., Partially supported by PID2021-122136OB-C21 by Ministerio de Ciencia e Innovación/AEI: (10.13039/501100011033/FEDER, UE), and 2021SGR00613-(ADBD) by Generalitat de Cataluña. The authors would like to thank Logroño City Hall for providing the access to the data., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

DragPoser: motion reconstruction from variable sparse tracking signals via latent space optimization

  • Pontón Martínez, José Luis
  • Pujol Puig, Eduard
  • Aristidou, Andreas
  • Andújar Gran, Carlos Antonio
  • Pelechano Gómez, Núria
High-quality motion reconstruction that follows the user's movements can be achieved by high-end mocap systems with many sensors. However, obtaining such animation quality with fewer input devices is gaining popularity as it brings mocap closer to the general public. The main challenges include the loss of end-effector accuracy in learning-based approaches, or the lack of naturalness and smoothness in IK-based solutions. In addition, such systems are often finely tuned to a specific number of trackers and are highly sensitive to missing data, e.g., in scenarios where a sensor is occluded or malfunctions. In response to these challenges, we introduce DragPoser, a novel deep-learning-based motion reconstruction system that accurately represents hard and dynamic constraints, attaining real-time high end-effectors position accuracy. This is achieved through a pose optimization process within a structured latent space. Our system requires only one-time training on a large human motion dataset, and then constraints can be dynamically defined as losses, while the pose is iteratively refined by computing the gradients of these losses within the latent space. To further enhance our approach, we incorporate a Temporal Predictor network, which employs a Transformer architecture to directly encode temporality within the latent space. This network ensures the pose optimization is confined to the manifold of valid poses and also leverages past pose data to predict temporally coherent poses. Results demonstrate that DragPoser surpasses both IK-based and the latest data-driven methods in achieving precise end-effector positioning, while it produces natural poses and temporally coherent motion. In addition, our system showcases robustness against on-the-fly constraint modifications, and exhibits adaptability to various input configurations and changes. The complete source code, trained model, animation databases, and supplementary material used in this paper can be found at https://upc-virvig.github.io/DragPoser, This work has received funding from the European Union’s Horizon 2020 Research and Innovation Programme under HORIZON-CL4-2022-HUMAN-01 grant agreement No 101093159 (XR4ED) and grant agreement No 739578 (RISE), from MCIN/AEI/10.13039/501100011033/FEDER “A way to make Europe”, UE (PID2021-122136OB-C21), from the Departament of Research and Universities of the Government of Catalonia (2021 SGR 01035), and the Government of the Republic of Cyprus through the Deputy Ministry of Research, Innovation and Digital Policy. Jose Luis Ponton was also funded by the Spanish Ministry of Universities (FPU21/01927), and Eduard Pujol by Universitat Politècnica de Catalunya and Banco Santander (FPI-UPC predoctoral grant)., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Simulating real-life scenarios to better understand the spread of diseases under different contexts

  • Blanco Guerra, Rafael Andrés
  • Patow, Gustavo Ariel
  • Pelechano Gómez, Núria
Current statistical models to simulate pandemics miss the most relevant information about the close atomic interactions between individuals which is the key aspect of virus spread. Thus, they lack a proper visualization of such interactions and their impact on virus spread. In the field of computer graphics, and more specifically in computer animation, there have been many crowd simulation models to populate virtual environments. However, the focus has typically been to simulate reasonable paths between random or semi-random locations in a map, without any possibility of analyzing specific individual behavior. We propose a crowd simulation framework to accurately simulate the interactions in a city environment at the individual level, with the purpose of recording and analyzing the spread of human diseases. By simulating the whereabouts of agents throughout the day by mimicking the actual activities of a population in their daily routines, we can accurately predict the location and duration of interactions between individuals, thus having a model that can reproduce the spread of the virus due to human-to-human contact. Our results show the potential of our framework to closely simulate the virus spread based on real agent-to-agent contacts. We believe that this could become a powerful tool for policymakers to make informed decisions in future pandemics and to better communicate the impact of such decisions to the general public., This work has received funding from the European Union’s Horizon 2020 research and innovation program under the Marie Skłodowska Curie grant agreement No 860768 (CLIPE project). Also, it was partially funded by grants PID2021-122136OB- C21 and PID2021-122136OB-C22 funded by MCIN/AEI/https://doi.org/10.13039/ 50110 0011033 and by ERDF A way of making Europe., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Seeing through others’ eyes in immersive VR: insights from a user study

  • Monclús Lahoya, Eva
  • Fairén González, Marta
  • Navazo Álvaro, Isabel
  • Vázquez Alcocer, Pere Pau
Despite the increasing usage of collaborative Virtual Reality in classrooms, teachers often struggle to ascertain whether the participants are adequately sharing the same viewpoint and are seeing the same information during lectures. Virtual Reality offers numerous advantages, including immersive experiences that enhance traditional classroom instruction and foster active and cooperative learning. A relevant part of the teaching process involves pointing to concrete regions or points of the virtual scene, and describing them. At this point, it is important that the pupils are able to see the same that is being seen by the instructor. In this paper we analyze viewpoint sharing methods commonly used in the literature, like teleport, 2D views with or without continuous streaming, and propose a novel method, dubbed Model Movement (MM). Our application layout is structured around a central 3D model that acts as the primary focal point at the collaborative session and the experiments were performed using 3D anatomical models. We conducted a user study with 32 participants (local and nominally remote) where participants were asked multiple times to swap to the instructor’s viewpoint and tracked their response times and their movement in the scene, among other variables. The data analysis and results of the questionnaires indicate that MM appears to be the preferred, more effective and usable of the analyzed techniques at showing the region of interest to inspect., This work has been partially supported by the project PID2021-122136OB-C21 funded by MCIN/AEI/ 10.13039/501100011033 and ERDF “A way of making Europe”, by the EU Horizon 2020 and the Department of Research and Universities of the Government of Catalonia (2021 SGR 01035), Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Virtual reality inspection of chromatin 3D and 2D data

  • Molina López, Elena
  • Kouril, David
  • Isenberg, Tobias
  • Kozliková, Barbora
  • Vázquez Alcocer, Pere Pau
Understanding the packing of long DNA strands into chromatin is one of the ultimate challenges in genomic research. An intrinsic part of this complex problem is studying the chromatin’s spatial structure. Biologists reconstruct 3D models of chromatin from experimental data, yet the exploration and analysis of such 3D structures is limited in existing genomic data visualization tools. To improve this situation, we investigated the current options of immersive methods and designed a prototypical VR visualization tool for 3D chromatin models that leverages virtual reality to deal with the spatial data. We showcase the tool in three primary use cases. First, we provide an overall 3D shape overview of the chromatin to facilitate the identification of regions of interest and the selection for further investigation. Second, we include the option to export the selected regions and elements in the BED format, which can be loaded into common analytical tools. Third, we integrate epigenetic modification data along the sequence that influence gene expression, either as in-world 2D charts or overlaid on the 3D structure itself. We developed our application in collaboration with two domain experts and gathered insights from two informal studies with five other experts., This project has been supported by PID2021-122136OB-C21 from the Ministerio de Ciencia e Innovación, by 839 FEDER (EU) funds, and The Czech Science Foundation research project no. GA23-05651S. Elena Molina has been supported by an FPU grant from the Ministerio de Ciencia e Innovación., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Terrain synthesis and authoring based on iso-contours

  • Huftier, Benoit
  • Schott, Hugo
  • Galin, Eric
  • Argudo Medrano, Óscar
  • Peytavie, Adrien
  • Guerin, Eric
Digital terrains are central to realistic landscape depiction, yet authoring tools must balance perceptual realism with intuitive artistic control. We propose a compact vector-based representation that models terrain as nested iso-contours, inspired by geomorphology and cartography. Our method departs from traditional grid-based elevation models by generating contours through an inward Open Eden Growth simulation, followed by marching-triangles reconstruction into a Triangulated Irregular Network. This contour framework supports direct editing such as warping, slope modulation, and smoothing, while allowing reconstruction of a standard elevation map for downstream processing, including erosion and amplification. The approach enables the creation of diverse, realistic terrains from minimal user input and offers simple yet powerful control for designers., This work is funded by the project EOLE ANR-23-CE56-0008, supported by Agence Nationale de la Recherche (France), and Grant PID2021-122136OB-C21 funded by MI-CIU/AEI/10.13039/501100011033 and ERDF/EU. Open access publication funding provided by COUPERIN CY26., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Towards the virtual reconstruction of physically lost altarpieces through differentiable rendering

  • Pagès Vilà, Anna
  • Muñoz Pandiella, Imanol
The interest of human beings in artistic creation, and its preservation, has been a constant throughout history. Thanks to this, today we can enjoy works from various moments and origins. However, not all works have been so fortunate, and many have disappeared for various reasons. To mitigate this loss, the virtual recreation of lost works of art is an open topic of interest for the computer graphics and cultural heritage communities. In the literature, we can find works along this line that are based on manual modeling or the use of numerous modern photographs. In this interdisciplinary work, however, we present a new methodology for reconstructing destroyed altarpieces for which only a single historical image is preserved, leveraging modern techniques of differentiable rendering. Additionally, we test our methodology in the reconstruction of the altarpiece of Sant Joan Baptista (Valls, Spain) destroyed during the Spanish Civil War. Our approach results in the partial 3D reconstruction of the main reliefs of the altarpiece. These results are truly useful, both for experts and for the public, as they allow for a better understanding of its volumetrics and its relationship with space, representing an important step forward in the virtual recovery of lost artifacts., This work has been partially supported by the projects PID2021-122136OB-C21 and PID2021-122136OB-C22 funded by MCIN/AEI/ 10.13039/501100011033 and by FEDER A way of making Europe, by the project PID2020- 114339GB-C21 funded by the MCIN/AEI, and by the Department of Research and Universities of the Generalitat de Catalunya (2021 SGR 01035)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Environment-aware motion matching

  • Pontón Martínez, José Luis
  • Andrews, Sheldon
  • Andújar Gran, Carlos Antonio
  • Pelechano Gómez, Núria
Interactive applications demand believable characters that respond naturally to dynamic environments. Traditional character animation techniques often struggle to handle arbitrary situations, leading to a growing trend of dynamically selecting motion-captured animations based on predefined features. While Motion Matching has proven effective for locomotion by aligning to target trajectories, animating environment interactions and crowd behaviors remains challenging due to the need to consider surrounding elements. Existing approaches often involve manual setup or lack the naturalism of motion capture. Furthermore, in crowd animation, body animation is frequently treated as a separate process from trajectory planning, leading to inconsistencies between body pose and root motion. To address these limitations, we present Environment-aware Motion Matching, a novel real-time system for full-body character animation that dynamically adapts to obstacles and other agents, emphasizing the bidirectional relationship between pose and trajectory. In a preprocessing step, we extract shape, pose, and trajectory features from a motion capture database. At runtime, we perform an efficient search that matches user input and current pose while penalizing collisions with a dynamic environment. Our method allows characters to naturally adjust their pose and trajectory to navigate crowded scenes., This work has received funding from MCIN/AEI/10.13039/501100011033/FEDER, UE (Spain) in the framework of the project PID2021-122136OB-C21, and with the support of the Department of Research and Universities of the Government of Catalonia (2021 SGR 01035). Jose Luis Ponton was funded by the Spanish Ministry of Universities (FPU21/01927). We thank the 2025 Bellairs Workshop on Computer Animation, where the initial ideas for this research took shape., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Automatic colon segmentation on T1-FS MR images

  • Orellana Bech, Bernat
  • Navazo Álvaro, Isabel
  • Brunet Crosa, Pere
  • Monclús Lahoya, Eva
  • Bendezú García, Álvaro
  • Azpiroz Vidaur, Fernando
The volume and distribution of the colonic contents provides valuable insights into the effects of diet on gut microbiotica involving both clinical diagnosis and research. In terms of Magnetic Resonance Imaging modalities, T2-weighted images allow the segmentation of the colon lumen, while fecal and gas contents can be only distinguished on the T1-weighted Fat-Sat modality. However, the manual segmentation of T1-weighted Fat-Sat is challenging, and no automatic segmentation methods are known. This paper proposed a non-supervised algorithm providing an accurate T1-weighted Fat-Sat colon segmentation via the registration of an existing colon segmentation in T2-weighted modality. The algorithm consists of two phases. It starts with a registration process based on a classical deformable registration method, followed by a novel Iterative Colon Registration process that utilizes a mesh deformation approach. This approach is guided by a probabilistic model that provides the likelihood of the colon boundary, followed by a shape preservation process of the colon segmentation on T2-weighted images. The iterative process converges to achieve an optimal fit for colon segmentation in T1-weighted Fat-Sat images. The segmentation algorithm has been tested on multiple datasets (154 scans) and acquisition machines (3) as part of the proof of concept for the proposed methodology. The quantitative evaluation was based on two metrics: the percentage of ground truth labeled feces correctly identified by our proposal (93+-5% ), and the volume variation between the existing colon segmentation in the T2-weighted modality and the colon segmentation computed in T1-weighted Fat-Sat images. Quantitative and medical evaluations demonstrated a degree of accuracy, usability, and stability concerning the acquisition hardware, making the algorithm suitable for clinical application and research., This work was supported in part by the projects PID2021-122295OB-I00 (Ministerio de Ciencia e Innovación, Spain), the project PID2021-122136OB-C21 funded by MCIN/AEI/ 10.13039/501100011033 and ERDF “A way of making Europe”, by the EU Horizon 2020 and the Department of Research and Universities of the Government of Catalonia (2021 SGR 01035). Ciberehd is funded by the Instituto de Salud Carlos III, Spain., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Simulation of mechanical weathering for modeling rocky terrains

  • Mateos Arlanzón, Diego
  • Carranza Cobeñas, Luis Alberto
  • Susín Sánchez, Antonio
  • Argudo Medrano, Óscar
Synthetic terrains play a vital role in various applications, including entertainment, training, and simulation. This work focuses on rocky terrains akin to those found in alpine environments, which contain many complex features such as sharp ridges, loose blocks, or overhangs that are often inadequately represented by standard 2D elevation maps. We propose a novel method based on a simplified simulation of mechanical erosion processes commonly observed in high-altitude terrains, in particular the weathering due to freeze-thaw cycles. The ultimate objective is to generate plausible rocky geometry from existing 3D models, as well as account for the temporal evolution due to these weathering processes. Additionally, we have developed an artist-friendly tool integrated as an add-on into Blender., This work has been funded with Grant PID2021-122136OB-C21 funded by MICIU/AEI/10.13039/501100011033 and ERDF/EU, and with the support of the Department of Research and Universi- ties of the Government of Catalonia (2021 SGR 01035). Luis Carranza was supported with Grant PRE2022-102912 funded by MI- CIU/AEI/10.13039/501100011033 and ESF+, and Oscar Argudo with a Maria Zambrano fellowship from Ministerio de Universidades and European Union—NextGenerationEU., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Are LLMs ready for visualization?

  • Vázquez Alcocer, Pere Pau
© 2024 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., Generative models have received a lot of attention in many areas of academia and the industry. Their capabilities span many areas, from the invention of images given a prompt to the generation of concrete code to solve a certain programming issue. These two paradigmatic cases fall within two distinct categories of requirements, ranging from "creativity" to "precision", as characterized by Bing Chat, which employs ChatGPT-4 as its backbone. Visualization practitioners and researchers have wondered to what end one of such systems could accomplish our work in a more efficient way. Several works in the literature have utilized them for the creation of visualizations. And some tools such as Lida, incorporate them as part of their pipeline. Nevertheless, to the authors’ knowledge, no systematic approach for testing their capabilities has been published, which includes both extensive and in-depth evaluation. Our goal is to fill that gap with a systematic approach that analyzes three elements: whether Large Language Models are capable of correctly generating a large variety of charts, what libraries they can deal with effectively, and how far we can go to configure individual charts. To achieve this objective, we initially selected a diverse set of charts, which are commonly utilized in data visualization. We then developed a set of generic prompts that could be used to generate them, and analyzed the performance of different LLMs and libraries. The results include both the set of prompts and the data sources, as well as an analysis of the performance with different configurations., Supported by Ministerio de Ciencia e Innovación/AEI (PID2021-122136OB-C21 by 10.13039/501100011033/FEDER, UE)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Rendering piecewise approximations of SDFs through analytic intersections

  • Pujol Puig, Eduard
  • Chica Calaf, Antonio
Signed distance fields (SDFs) have emerged as an alternative shape representation for real-time collision detection and lighting effects. Computing these for complex models can be expensive, so one popular approach is to prepare an approximation via sampling and interpolation. Then, these may be rendered using sphere marching, which gets close to the surface quickly, but needs several iterations to converge to it. In this paper, we propose an alternative that computes the intersection of a given ray and the surface analytically at a narrow band. This may be combined with other enhancements like having variable error for the approximation depending on the distance to the surface and skipping regions that do not contain the surface to accelerate the outer band ray traversal while reducing the required memory. To achieve smoother representations with minimal computational cost, we propose a method for computing surface intersections and normals from separate interpolants. We evaluate all these to find the optimal combination improving the rendering performance and memory consumption of these SDF approximations., This work has been partially funded by Ministeri de Ciència i Innovació(MICIN), Agencia Estatal de Investigación (AEI) and the Fons Europeu de Desenvolupament Regional (FEDER) (project PID2021-122136OB-C21 funded by MCIN/AEI/10.13039/501100011033/ FEDER, UE); and by the Department of Research and Universities of the Government of Catalonia (2021 SGR 01035). The first author gratefully acknowledges the Universitat Politècnica de Catalunya and Banco Santander for the financial support of his predoctoral grant FPI UPC grant., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

State of the art of molecular visualization in immersive virtual environments

  • Kuták, David
  • Vázquez Alcocer, Pere Pau
  • Isenberg, Tobias
  • Krone, Michael
  • Baaden, Marc
  • Byska, Jan
  • Kozliková, Barbora
  • Miao, Haichao
Visualization plays a crucial role in molecular and structural biology. It has been successfully applied to a variety of tasks, including structural analysis and interactive drug design. While some of the challenges in this area can be overcome with more advanced visualization and interaction techniques, others are challenging primarily due to the limitations of the hardware devices used to interact with the visualized content. Consequently, visualization researchers are increasingly trying to take advantage of new technologies to facilitate the work of domain scientists. Some typical problems associated with classic 2D interfaces, such as regular desktop computers, are a lack of natural spatial understanding and interaction, and a limited field of view. These problems could be solved by immersive virtual environments and corresponding hardware, such as virtual reality head-mounted displays. Thus, researchers are investigating the potential of immersive virtual environments in the field of molecular visualization. There is already a body of work ranging from educational approaches to protein visualization to applications for collaborative drug design. This review focuses on molecular visualization in immersive virtual environments as a whole, aiming to cover this area comprehensively. We divide the existing papers into different groups based on their application areas, and types of tasks performed. Furthermore, we also include a list of available software tools. We conclude the report with a discussion of potential future research on molecular visualization in immersive environments., This work was supported by the US DOE LLNL-LDRD 22-FS- 026 and was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344 (LLNL-CONF-831245) and by project TIN2017-88515-C2-1-R (GEN3DLIVE), from the Spanish Ministerio de Economía y Competitividad, by 839 FEDER (EU) funds. The work received further support by the French National Research agency (Grant “PIRATE”, ANR-21-CE45-0014) and by the “Initiative d’Excellence” program from the French State (Grant “DYNAMO”, ANR-11-LABX-0011 and grant “CACSICE”, ANR-11-EQPX-0008), and from Carl Zeiss Stiftung. Ministerio de Ciencia e Innovación/AEI (PID2021-122136OB- C21[10.13039/501100011033/FEDER, UE]), Peer Reviewed, Postprint (published version)