Skip to main content

RECONSTRUCCION DE OBJETOS DEFORMABLES A PARTIR DE IMAGENES Y SUS APLICACIONES A LA REALIDAD AUMENTADA EN CIRUGIA MINIMAMENTE INVASIVA

TIN2016-80939-R

•
Funding agency name Agencia Estatal de Investigación
Funding agency acronym AEI
Program Programa Estatal de I+D+i Orientado a los Retos de la Sociedad
Call Proyectos I+D+i Retos de Investigación
Call year 2016
State Plan 2013-2016 Plan Estatal de Investigación Científica y Técnica y de Innovación
Beneficiary centre UNIVERSIDAD DE ALCALA

Resultados relacionados

Resultados totales (Incluyendo duplicados): 15
Encontrada(s) 1 página(s)

DIGITAL.CSIC. Repositorio Institucional del CSIC
Publication Artículo científico (JournalArticle)

A multi-poosition approach in a smart fiber-optic surveillance system for pipeline integrity threat detection

  • Tejedor, Javier
  • Macías-Guarasa, Javier
  • Martins, Hugo F.
  • Martín-López, Sonia
  • González-Herráez, Miguel
19 pags., 4 figs., 9 tabs. -- This article belongs to the Special Issue Pattern Recognition and Applications, We present a new pipeline integrity surveillance system for long gas pipeline threat detection and classification. The system is based on distributed acoustic sensing with phase-sensitive optical time domain reflectometry (ϕ-OTDR) and pattern recognition for event classification. The proposal incorporates a multi-position approach in a Gaussian Mixture Model (GMM)-based pattern classification system which operates in a real-field scenario with a thorough experimental procedure. The objective is exploiting the availability of vibration-related data at positions nearby the one actually producing the main disturbance to improve the robustness of the trained models. The system integrates two classification tasks: (1) machine + activity identification, which identifies the machine that is working over the pipeline along with the activity being carried out, and (2) threat detection, which aims to detect suspicious threats for the pipeline integrity (independently of the activity being carried out). For the machine + activity identification mode, the multi-position approach for model training obtains better performance than the previously presented single-position approach for activities that show consistent behavior and high energy (between 6% and 11% absolute) with an overall increase of 3% absolute in the classification accuracy. For the threat detection mode, the proposed approach gets an 8% absolute reduction in the false alarm rate with an overall increase of 4.5% absolute in the classification accuracy., This research was funded by a private research project by Fluxys, Gassco and Statoil named “PIT-STOP (Early detection of Pipeline Integrity Threats using a SmarT fiber-OPtic surveillance system)”, by the Spanish Ministry of Economy and Competitiveness under projects HEIMDALUAH (TIN2016-75982-C2-1-R) and ARTEMISA (TIN2016-80939-R), by the University of Alcalá under projects ACUFANO (CCG19/IA-024) and ARGOS (CCG20/IA-043), and by the Spanish Ministry of Science, Innovation and Universities (IJCI-2017-33856 and RTI2018-097957-B-C33). The APC was funded by the Ministry of Science, Innovation and Universities of Spain grant number RTI2018-095324-B-I00., Peer reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publication Artículo científico (JournalArticle)

A hybrid cascade-parallel discriminative-generative model for pipeline integrity threat detection in a smart fiber optic surveillance system

  • Tejedor, Javier
  • Macias-Guarasa, Javier
  • Martins, Hugo F.
  • Martín-López, Sonia
  • González-Herráez, Miguel
25 pags., 10 figs., 8 tabs., This paper presents an advanced system for the continuous monitoring of potential threats in a long gas pipeline. For signal acquisition, phase-sensitive optical time domain reflectometry (ϕ-OTDR) technology is employed. Then, pattern recognition strategies are incorporated, which are aimed at identifying threats. To do so, the system integrates a random forest-based approach on top of a multiple-layer perceptron (MLP)-based discriminative approach for feature extraction within a parallel Gaussian Mixture Model (GMM)-Hidden Markov Model (HMM) for pattern classification in a hybrid approach. Subsequently, a system combination strategy, which makes use of the decisions carried out by this hybrid approach, is also presented. This strategy is based on the so-called majority voting technique, which makes use of the output of the classification step from the different feature extraction strategies and the different number of states in the GMM-HMM-based classification. The system is tested on two tasks: (1) Identification of machine and activity, and (2) detection of threats for the pipeline. Compared with our previous system, the results of this advanced system show that the hybrid feature extraction and pattern classification achieve statistically significant improvements for both tasks (i.e., 5% of relative improvement for the machine and activity identification task, 1% of relative improvement in the threat detection rate, and 15% of relative improvement in the false alarm rate for the threat detection task)., This work was partially supported by the Ministry of Science, Innovation and Universities of Spain (grant number RTI2018-095324-B-I00).This work was also funded by the Spanish Ministry of Economy and Competitiveness with projects ARTEMISA (TIN2016-80939-R) and HEIMDAL-UAH (TIN201675982-C2-1-R), by the Spanish Ministry of Science and Innovation MCIN/AEI/10.13039/501100011033 and by the European Union NextGenerationEU/PRTR program, with projects PSI (PLEC2021-007875),ATHENA (PID2020-115995RB-I00) and EYEFUL (PID2020-113118RB-C31), and by CAM and UAH under projects ARGOS+ (PIUAH21/IA-016) and CONDORDIA (CM/JIN/2021-015). The authors gratefully acknowledge the computer resources at Artemisa, funded by the European Union ERDF and Comunitat Valenciana as well as the technical support provided by the Instituto de Fisica Corpuscular, IFIC (CSIC-UV). The authors thank Sira E. Palazuelos-Cagigas for her participation in the writing and technical editing of the manuscript., Peer reviewed
e_Buah Biblioteca Digital Universidad de Alcalá
Publication Artículo científico (JournalArticle)

Semi-supervised anomaly detection in video-surveillance scenes in the wild

  • Sarker, Mohammad Ibrahim
  • Losada Gutiérrez, Cristina|||0000-0001-9545-327X
  • Marrón Romera, Marta|||0000-0001-7723-2262
  • Fuentes Jiménez, David|||0000-0001-6424-4782
  • Luengo Sánchez, Sara|||0000-0003-3942-3804
Surveillance cameras are being installed in many primary daily living places to maintain public safety. In this video-surveillance context, anomalies occur only for a very short time, and very occasionally. Hence, manual monitoring of such anomalies may be exhaustive and monotonous, resulting in a decrease in reliability and speed in emergency situations due to monitor tiredness. Within this framework, the importance of automatic detection of anomalies is clear, and, therefore, an important amount of research works have been made lately in this topic. According to these earlier studies, supervised approaches perform better than unsupervised ones. However, supervised approaches demand manual annotation, making dependent the system reliability of the different situations used in the training (something difficult to set in anomaly context). In this work, it is proposed an approach for anomaly detection in video-surveillance scenes based on a weakly supervised learning algorithm. Spatio-temporal features are extracted from each surveillance video using a temporal convolutional 3D neural network (T-C3D). Then, a novel ranking loss function increases the distance between the classification scores of anomalous and normal videos, reducing the number of false negatives. The proposal has been evaluated and compared against state-of-art approaches, obtaining competitive performance without fine-tuning, which also validates its generalization capability. In this paper, the proposal design and reliability is presented and analyzed, as well as the aforementioned quantitative and qualitative evaluation in-the-wild scenarios, demonstrating its high sensitivity in anomaly detection in all of them., info:eu-repo/grantAgreement/UAH//CCG2020%2FIA-043, Agencia Estatal de Investigación, Ministerio de Economía y Competitividad, Universidad de Alcalá
e_Buah Biblioteca Digital Universidad de Alcalá
Publication Artículo científico (JournalArticle)

A multi-position approach in a smart fiber-optic surveillance system for pipeline integrity threat detection

  • Tejedor Noguerales, Javier
  • Macías Guarasa, Javier|||0000-0002-3303-3963
  • Fidalgo Martins, Hugo|||0000-0003-3927-8125
  • Martín López, Sonia|||0000-0001-5203-6206
  • González Herráez, Miguel|||0000-0003-2555-2971
We present a new pipeline integrity surveillance system for long gas pipeline threat detection and classification. The system is based on distributed acoustic sensing with phase-sensitive optical time domain reflectometry (?-OTDR) and pattern recognition for event classification. The proposal incorporates a multi-position approach in a Gaussian Mixture Model (GMM)-based pattern classification system which operates in a real-field scenario with a thorough experimental procedure. The objective is exploiting the availability of vibration-related data at positions nearby the one actually producing the main disturbance to improve the robustness of the trained models. The system integrates two classification tasks: (1) machine + activity identification, which identifies the machine that is working over the pipeline along with the activity being carried out, and (2) threat detection, which aims to detect suspicious threats for the pipeline integrity (independently of the activity being carried out). For the machine + activity identification mode, the multi-position approach for model training obtains better performance than the previously presented single-position approach for activities that show consistent behavior and high energy (between 6% and 11% absolute) with an overall increase of 3% absolute in the classification accuracy. For the threat detection mode, the proposed approach gets an 8% absolute reduction in the false alarm rate with an overall increase of 4.5% absolute in the classification accuracy., Ministerio de Economía y Competitividad, Agencia Estatal de Investigación, Universidad de Alcalá, Ministerio de Ciencia e Innovación
e_Buah Biblioteca Digital Universidad de Alcalá
Publication Artículo científico (JournalArticle)

Towards dense people detection with deep learning and depth images

  • Fuentes Jiménez, David|||0000-0001-6424-4782
  • Losada Gutiérrez, Cristina|||0000-0001-9545-327X
  • Casillas Pérez, David|||0000-0002-5721-1242
  • Macías Guarasa, Javier|||0000-0002-3303-3963
  • Pizarro Pérez, Daniel|||0000-0003-0622-4884
  • Martín López, Roberto|||0000-0002-5172-9403
  • Luna Vázquez, Carlos Andrés
This paper describes a novel DNN-based system, named PD3net, that detects multiple people from a single depth image, in real time. The proposed neural network processes a depth image and outputs a likelihood map in image coordinates, where each detection corresponds to a Gaussian-shaped local distribution, centered at each person?s head. This likelihood map encodes both the number of detected people as well as their position in the image, from which the 3D position can be computed. The proposed DNN includes spatially separated convolutions to increase performance, and runs in real-time with low budget GPUs. We use synthetic data for initially training the network, followed by fine tuning with a small amount of real data. This allows adapting the network to different scenarios without needing large and manually labeled image datasets. Due to that, the people detection system presented in this paper has numerous potential applications in different fields, such as capacity control, automatic video-surveillance, people or groups behavior analysis, healthcare or monitoring and assistance of elderly people in ambient assisted living environments. In addition, the use of depth information does not allow recognizing the identity of people in the scene, thus enabling their detection while preserving their privacy. The proposed DNN has been experimentally evaluated and compared with other state-of-the-art approaches, including both classical and DNN-based solutions, under a wide range of experimental conditions. The achieved results allows concluding that the proposed architecture and the training strategy are effective, and the network generalize to work with scenes different from those used during training. We also demonstrate that our proposal outperforms existing methods and can accurately detect people in scenes with significant occlusions., Ministerio de Economía y Competitividad, Universidad de Alcalá, Agencia Estatal de Investigación
e_Buah Biblioteca Digital Universidad de Alcalá
Publication Artículo científico (JournalArticle)

Equiareal Shape-from-Template

  • Casillas Pérez, David|||0000-0002-5721-1242
  • Pizarro Pérez, Daniel|||0000-0003-0622-4884
  • Fuentes Jiménez, David|||0000-0001-6424-4782
  • Mazo Quintas, Manuel Ramón
  • Bartoli, Adrien
This paper studies the 3D reconstruction of a deformable surface from a single image and a reference surface, known as the template. This problem is known as Shape-from-Template and has been recently shown to be well-posed for isometric deformations, for which the surface bends without altering geodesics. This pa- per studies the case of equiareal deformations. They are elastic deformations where the local area is pre- served and thus include isometry as a special case. Elas- tic deformations have been studied before in Shape- from-Template, yet no theoretical results were given on the existence or uniqueness of solutions. The equiareal model is much more widely applicable than the isomet- ric and the conformal models. This paper introduces Monge?s theory, widely used for studying the solutions of non-linear first-order PDEs to the field of 3D recon- struction. It uses this theory to establish a theoretical framework for equiareal Shape-from-Template and an- swers the important question of whether it is possible to reconstruct a surface exactly which a much weaker prior than isometry. Indead, we prove that equiareal Shape-from-Template has a maximum of two local so- lutions sufficiently near an initial curve that lies on the surface. In addition we propose a convex reconstruction algorithm that can recover the multiple solutions. Our algorithm uses standard numerical tools for ODEs. We use the perspective camera model and give reconstruc- tion results with both synthetic and real examples., Agencia Estatal de Investigación, Universidad de Alcalá, European Commission
e_Buah Biblioteca Digital Universidad de Alcalá
Publication Artículo científico (JournalArticle)

People re-identification using depth and intensity information from an overhead sensor

  • Luna Vázquez, Carlos Andrés
  • Losada Gutiérrez, Cristina|||0000-0001-9545-327X
  • Fuentes Jiménez, David|||0000-0001-6424-4782
  • Mazo Quintas, Manuel Ramón
This work presents a new people re-identification method, using depth and intensity images, both of them captured with a single static camera, located in an overhead position. The proposed solution arises from the need that exists in many areas of application to carry out identification and re-identification processes to determine, for example, the time that people remain in a certain space, while fulfilling the requirement of preserving people's privacy. This work is a novelty compared to other previous solutions, since the use of top-view images of depth and intensity allows obtaining information to perform the functions of identification and re-identification of people, maintaining their privacy and reducing occlusions. In the procedure of people identification and re-identification, only three frames of intensity and depth are used, so that the first one is obtained when the person enters the scene (frontal view), the second when it is in the central area of the scene (overhead view) and the third one when it leaves the scene (back view). In the implemented method only information from the head and shoulders of people with these three different perspectives is used. From these views three feature vectors are obtained in a simple way, two of them related to depth information and the other one related to intensity data. This increases the robustness of the method against lighting changes. The proposal has been evaluated in two different datasets and compared to other state-of-the-art proposal. The obtained results show a 96,7% success rate in re-identification, with sensors that use different operating principles, all of them obtaining depth and intensity information. Furthermore, the implemented method can work in real time on a PC, without using a GPU., Ministerio de Economía y Competitividad, Agencia Estatal de Investigación, Universidad de Alcalá
e_Buah Biblioteca Digital Universidad de Alcalá
Publication Artículo científico (JournalArticle)

DPDnet: a robust people detector using deep learning with an overhead depth camera

  • Fuentes Jiménez, David|||0000-0001-6424-4782
  • Martín López, Roberto|||0000-0002-5172-9403
  • Losada Gutiérrez, Cristina|||0000-0001-9545-327X
  • Casillas Pérez, David|||0000-0002-5721-1242
  • Macías Guarasa, Javier|||0000-0002-3303-3963
  • Luna Vázquez, Carlos Andrés
  • Pizarro Pérez, Daniel|||0000-0003-0622-4884
This paper proposes a deep learning-based method to detect multiple people from a single overhead depth image with high precision. Our neural network, called DPDnet, is composed by two fully-convolutional encoder-decoder blocks built with residual layers. The main block takes a depth image as input and generates a pixel-wise confidence map, where each detected person in the image is represented by a Gaussian-like distribution, The refinement block combines the depth image and the output from the main block, to refine the confidence map. Both blocks are simultaneously trained end-to-end using depth images and ground truth head position labels. The paper provides a rigorous experimental comparison with some of the best methods of the state-of-the-art, being exhaustively evaluated in different publicly available datasets. DPDnet proves to outperform all the evaluated methods with statistically significant differences, and with accuracies that exceed 99%. The system was trained on one of the datasets (generated by the authors and available to the scientific community) and evaluated in the others without retraining, proving also to achieve high accuracy with varying datasets and experimental conditions. Additionally, we made a comparison of our proposal with other CNN-based alternatives that have been very recently proposed in the literature, obtaining again very high performance. Finally, the computational complexity of our proposal is shown to be independent of the number of users in the scene and runs in real time using conventional GPUs., Ministerio de Economía y Competitividad, Universidad de Alcalá, Agencia Estatal de Investigación
e_Buah Biblioteca Digital Universidad de Alcalá
Publication Artículo científico (JournalArticle)

Towards End-to-End Acoustic Localization Using Deep Learning: From Audio Signals to Source Position Coordinates

  • Vera Díaz, Juan Manuel|||0000-0002-6152-5789
  • Pizarro Pérez, Daniel|||0000-0003-0622-4884
  • Macías Guarasa, Javier|||0000-0002-3303-3963
This paper presents a novel approach for indoor acoustic source localization using microphone arrays, based on a Convolutional Neural Network (CNN). In the proposed solution, the CNN is designed to directly estimate the three-dimensional position of a single acoustic source using the raw audio signal as the input information and avoiding the use of hand-crafted audio features. Given the limited amount of available localization data, we propose, in this paper, a training strategy based on two steps. We first train our network using semi-synthetic data generated from close talk speech recordings. We simulate the time delays and distortion suffered in the signal that propagate from the source to the array of microphones. We then fine tune this network using a small amount of real data. Our experimental results, evaluated on a publicly available dataset recorded in a real room, show that this approach is able to produce networks that significantly improve existing localization methods based on SRP-PHAT strategies and also those presented in very recent proposals based on Convolutional Recurrent Neural Networks (CRNN). In addition, our experiments show that the performance of our CNN method does not show a relevant dependency on the speaker's gender, nor on the size of the signal window being used., Universidad de Alcalá, Ministerio de Economía, Comercio y Empresa
e_Buah Biblioteca Digital Universidad de Alcalá
Publication Artículo científico (JournalArticle)

Label augmentation to improve generalization of deep learning semantic segmentation of laparoscopic images

  • Monasterio Expósito, Leticia
  • Pizarro Pérez, Daniel|||0000-0003-0622-4884
  • Macías Guarasa, Javier|||0000-0002-3303-3963
Training Deep Neural Networks to solve semantic segmentation is a challenging problem with small-size labeled datasets, leading to overfitting. This is especially problematic in medical images, and in particular, in laparoscopic surgery images. In this context, ground-truth segmentation labels are available only for a small set of images with few patients. Besides, inter-patient variability is very high in practice. Models trained for a specific setup and a set of patients usually performs poorly when deployed in a new environment. This work proposes a new training strategy that improves the generalization accuracy of current state-of-the-art semantic segmentation methods applied to laparoscopic images. Our approach is based on training a discriminator network, which learns to detect segmentation errors, producing a dense segmentation error map. Unlike in adversarial networks, we train the discriminator offline by synthetically altering ground- truth segmentation labels with simple morphological and geometric operations. We then use the discriminator to train a segmentation neural network, by minimizing the discriminator predicted error jointly with a standard segmentation loss. This strategy results in segmentation models that are significantly more accurate when tested in unseen images than those only relying on data augmentation. This technique is very suitable to boost the performance of any state-of-the-art segmentation network and can be combined with other data augmentation strategies. This paper evaluates and validates our proposal by training and testing common state-of-the-art segmentation models in publicly available semantic segmentation datasets, specialized in laparoscopic and endoscopic surgery. The results show that our methods are effective, obtaining a significant improvement in terms of segmentation accuracy, especially in challenging small-size datasets., Agencia Estatal de Investigación, Ministerio de Economía y Competitividad, Universidad de Alcalá
e_Buah Biblioteca Digital Universidad de Alcalá
Publication Artículo científico (JournalArticle)

Segregating the true perturbation position from ghost energy points region in ϕ-OTDR systems

  • Adeel, Muhammad
  • Tejedor Noguerales, Javier
  • Macías Guarasa, Javier|||0000-0002-3303-3963
  • Shang, Chao
  • Lu, Chao
ϕ-OTDR perturbation detection applications demand optimal precision of the perturbation location. Strategies for improving both signal-to-noise (SNR) and precision of the perturbation location in a laboratory environment may fail when applying to a very long fiber under test (FUT) in real-field environments. With this deployment, meaningful energy points representing the response of a certain perturbation can be located at random locations of the fiber other than the original location of the perturbation. These random locations are referred to as the ghost energy points that confuse the system to mistakenly consider the location of these points as the original perturbation location. We present in this paper a novel space-time scanning (ST-scan) method that segregates the ghost energy point locations from those of the real perturbation so that the original perturbation location estimation is improved., Agencia Estatal de Investigación, Ministerio de Economía y Competitividad, Universidad de Alcalá
e_Buah Biblioteca Digital Universidad de Alcalá
Publication Artículo científico (JournalArticle)

The Isowarp: The Template-Based Visual Geometry of Isometric Surfaces

  • Casillas Pérez, David|||0000-0002-5721-1242
  • Pizarro Pérez, Daniel|||0000-0003-0622-4884
  • Fuentes Jiménez, David|||0000-0001-6424-4782
  • Mozo Quintas, Manuel Ramón
  • Bartoli, Adrien
Registration maps or warps form a key element in Shape-from-Template (SfT). They relate the template with the input image, which contains the projection of the deformed surface. Recently, it was shown that isometric SfT can be solved analytically if the warp and its first-order derivatives are known. In practice, the warp is recovered by interpolating a set of discrete template-to-image point correspondences. This process relies on smoothness priors but ignores the 3D geometry. This may produce errors in the warp and poor reconstructions. In contrast, we propose to create a 3D consistent warp, which technically is a very challenging task, as the 3D shape variables must be eliminated from the isometric SfT equations to find differential constraints for the warp only. Integrating these constraints in warp estimation yields the isowarp, a warp 3D consistent with isometric SfT. Experimental results show that incorporating the isowarp in the SfT pipeline allows the analytic solution to outperform non-convex 3D shape refinement methods and the recent DNN-based SfT methods. The isowarp can be properly initialized with convex methods and its hyperparameters can be automatically obtained with cross-validation. The isowarp is resistant to 3D ambiguities and less computationally expensive than existing 3D shape refinement methods. The isowarp is thus a theoretical and practical breakthrough in SfT., Agencia Estatal de Investigación, European Commission
e_Buah Biblioteca Digital Universidad de Alcalá
Publication Artículo científico (JournalArticle)

Undiscovered issues and solutions for direct detected ϕ-OTDR systems

  • Adeel, Muhammad
  • Tejedor Noguerales, Javier
  • Macías Guarasa, Javier|||0000-0002-3303-3963
  • Zhu, Kun
Time and spatial domains in ϕ-OTDR perturbation detection and recognition for pipeline and border security applications in very long fiber under test (FUT) environments have not been properly analyzed so far. We propose in this pa- per several issues and the corresponding solutions that should be considered in both domains when developing those applications. Solutions are based on: (1) the importance of differential signals in direct detected ϕ-OTDR systems; (2) dealing with the useless data acquired in the time domain; (3) dealing with the zero and ghost energy points along the FUT which will undoubtedly lead to hiding the real perturbation and reducing the potential of finding the exact location of the perturbation, respectively. Experimental results show that: (1) setting the appropriate experimental conditions (i.e., increasing the intensity of the perturbation and sampling at the minimum sampling rate that satisfies the Nyquist criterion) is a must; (2) the Kendall correlation-based technique on differential signals allows obtaining more significant data traces both in time and spatial domains; (3) combining both methods obtains further gains; (4) S-Scan algorithm accurately provides the time instant of the real perturbation by increasing the probability of detecting zero energy points and avoiding ghost energy points in further processing., Ministerio de Economía y Competitividad, Agencia Estatal de Investigación, Universidad de Alcalá
e_Buah Biblioteca Digital Universidad de Alcalá
Publication Artículo científico (JournalArticle)

Model-based real-time non-rigid tracking

  • Bronte Palacios, Sebastián
  • Bergasa Pascual, Luis Miguel|||0000-0002-0087-3077
  • Pizarro Pérez, Daniel|||0000-0003-0622-4884
  • Barea Navarro, Rafael|||0000-0002-4179-6100
This paper presents a sequential non-rigid reconstruction method that recovers the 3D shape and the camera pose of a deforming object from a video sequence and a previous shape model of the object. We take PTAM (Parallel Mapping and Tracking), a state-of-the-art sequential real-time SfM (Structure-from-Motion) engine, and we upgrade it to solve non-rigid reconstruction. Our method provides a good trade-off between processing time and reconstruction error without the need for specific processing hardware, such as GPUs. We improve the original PTAM matching by using descriptor-based features, as well as smoothness priors to better constrain the 3D error. This paper works with perspective projection and deals with outliers and missing data. We evaluate the tracking algorithm performance through different tests over several datasets of non-rigid deforming objects. Our method achieves state-of-the-art accuracy and can be used as a real-time method suitable for being embedded in portable devices., Ministerio de Economía y Competitividad, Comunidad de Madrid, Agencia Estatal de Investigación
e_Buah Biblioteca Digital Universidad de Alcalá
Publication Artículo científico (JournalArticle)

Acoustic source localization with deep generalized cross correlations

  • Vera Díaz, Juan Manuel|||0000-0002-6152-5789
  • Pizarro Pérez, Daniel|||0000-0003-0622-4884
  • Macías Guarasa, Javier|||0000-0002-3303-3963
One of the most popular techniques for Acoustic Source Localization is the Generalized Cross Correlation (GCC) and its use in Steered Response Power techniques (SRP). Nowadays, Deep Learning strategies may outperform these classical methods, but they are generally dependent on the room and sensor geometric configuration that are used during the training phases. Hence, they require adaptation and re-training when facing a new environment, which is a problem in practice as re-training requires labelling new data and running a complex training algorithm. In this work we use a Convolutional Deep Neural Network that transforms the GCC between two signals into a Gaussian shaped signal, that we call Deep Generalized Cross Correlation (DeepGCC). We combine DeepGCC estimations to create a 3D acoustic map, similarly to SRP techniques. This acoustic map can be further refined using a sparse generative model to recover the source position. Crucially, we can adapt the acoustic map to different microphone array geometries without retraining the DeepGCC network. We show that our method outperforms both classical approaches and recent Deep Learning strategies in real and simulated challenging scenarios with mismatched training-testing conditions, not requiring re-training with different sensor configurations or room environments., Ministerio de Economía y Competitividad, Agencia Estatal de Investigación, Universidad de Alcalá