Skip to main content

RETORNO AL HOGAR: SISTEMA DE MEJORA DE LA AUTONOMIA DE PERSONAS CON DAÑO CEREBRAL ADQUIRIDO Y DEPENDIENTES EN SU INTEGRACION EN LA SOCIEDAD

TIN2016-76515-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 ALICANTE

Resultados relacionados

Resultados totales (Incluyendo duplicados): 18
Encontrada(s) 1 página(s)

BULERIA. Repositorio Institucional de la Universidad de León
Publication Artículo científico (JournalArticle)

A context-awareness model for activity recognition in robot-assisted scenarios

  • Rodríguez Lera, Francisco Javier
  • Martín Rico, Francisco
  • Guerrero Higueras, Ángel Manuel
  • Matellán Olivera, Vicente
[ES] La conciencia del contexto en los programas de vida asistida para personas mayores es una piedra angular en el escenario actual de robots de servicio no personalizados distribuidos por todo el mundo. Esta investigación propone un sistema de conciencia del contexto para la interpretación de una escena humano-robot basado en siete contextos primarios y en la Asociación Americana de Terapia Ocupacional. El sistema de conocimiento del contexto aquí definido propone un mecanismo de inferencia para el reconocimiento de la actividad basado en redes bayesianas jerárquicas. Sin embargo, cuando aumenta la información procedente de los sensores, también aumenta el coste computacional asociado. Por ello, es necesaria una evaluación de diferentes modelos de redes bayesianas para disminuir su impacto sobre el rendimiento del robot. Se han modelado y probado dos modelos topológicos utilizando la aplicación OpenMarkov: un enfoque de dos niveles de una capa de entrada-observaciones y la capa de reconocimiento de la actividad, y un modelo de tres capas que separa una capa de contextos primarios, la capa de entrada-observaciones y la capa de reconocimiento de la actividad. Los resultados cualitativos y cuantitativos presentados muestran un mejor rendimiento en términos de memoria y memoria en un modelo de tres capas. Además, se presenta su efecto en una arquitectura híbrida de una plataforma robótica., [EN] Context awareness in ambient assisted living programmes for the elderly is a cornerstone in the current scenario of noncustomized service robots distributed around the world. This research proposes a context-awareness system for a human–robot scene interpretation based on seven primary contexts and the American Occupational Therapy Association. The context-awareness system defined here proposes an inference mechanism for the activity recognition supported on hierarchical Bayesian networks. However, when the information from sensors increases, the computational cost associated also increases. Thus, an evaluation of different Bayesian network models is necessary for decreasing its impact over the robot performance. Two topological models have been modelled and tested using OpenMarkov application: a two-level approach of an input–observations layer and the activity recognition layer, and a three-layer model setting apart a primary contexts layer, the input–observations layer, and the activity recognition layer. The qualitative and quantitative results presented here show better performance in terms of memory and memory in a three-layer model. Besides, its effect on a hybrid architecture of a robotic platform is presented., SI, Junta de Castilla y León and FEDER funds, Spain, under Research Grant LE028P17
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

Entorno Docente Universitario para la Programación de los Robots, Academic framework for teaching robot programming at university

  • Cañas Plaza, José María
  • Martín Florido, Alberto
  • Perdices García, Eduardo
  • Rivas Montero, Francisco
  • Calvo Palomino, Roberto
[ES] Se presenta un entorno docente de robótica universitaria orientado al aprendizaje práctico. Es software libre, multiplataforma (Linux, Windows, MacOS) y hace énfasis en la programación de la inteligencia de los robots. Consta de una colección de prácticas variadas, cercanas a las aplicaciones robóticas que están llegando a la sociedad recientemente (coches autónomos, drones, aspiradoras...).Utiliza el simulador Gazebo como referencia y Python como lenguaje. Para cada práctica se ha programado una aplicación académica que realiza tareas auxiliares como la interfaz gráfica, la conexión con sensores y actuadores concretos, la temporización, etc. y aloja al código del estudiante, que así se concentra en los algoritmos de percepción y control. Cada aplicación está formada por una parte específica preparada, que queda oculta, y el código del estudiante, que simplemente rellena un sencillo fichero plantilla con la lógica del robot. Se ha utilizado con éxito en varios cursos de grado, de máster y cursos de introducción a la robótica., [EN] This paper presents a framework for teaching robotics at engineering university degrees in a practical way. It is open source, multiplatform (Linux, Windows, MacOS) and emphasizes the programming of the robot intelligence. It consists of a collection of exercises of several types, which are similar to the new robotic applications (autonomous vehicles, drones, vacuum cleaners...). It uses the Gazebo simulator and the Python programming language. For each exercise an academic application has been created which performs all the auxiliary tasks like graphical interface, connection to the sensors and actuators, timing of the code, etc. and hosts the student’s code, who can focus on the perception and control algorithms. Each academic application combines an infrastructure part, which it is provided and remains hidden, and the student part, who simply fills a form file with the robot logic for such exercise. It has been successfully used in several subjects at engineering degree, master and robotics introductory courses., Este trabajo ha sido financiado parcialmente por la Comunidad de Madrid a través del proyecto RoboCity2030-III ((S2013/MIT-2748) y por el Ministerio de Economía y Competitividad de España a través del proyecto RETOGAR (TIN2016-76515-R). Los autores agradecen a Google la financiación al proyecto JdeRobot en sus convocatorias Google Summer of Code de 2015 y 2017.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

PHAROS 2.0-A PHysical Assistant RObot System Improved

  • Martinez-Martin, Ester
  • Araujo, Angelo
  • Cazorla, Miguel
[EN] There are great physical and cognitive benefits for older adults who are engaged in active aging, a process that should involve daily exercise. In our previous work on the PHysical Assistant RObot System (PHAROS), we developed a system that proposed and monitored physical activities. The system used a social robot to analyse, by means of computer vision, the exercise a person was doing. Then, a recommender system analysed the exercise performed and indicated what exercise to perform next. However, the system needed certain improvements. On the one hand, the vision system captured the movement of the person and indicated whether the exercise had been done correctly or not. On the other hand, the recommender system was based purely on a ranking system that did not take into account temporal evolution and preferences. In this work, we propose an evolution of PHAROS, PHAROS 2.0, incorporating improvements in both of the previously mentioned aspects. In the motion capture aspect, we are now able to indicate the degree of completeness of each exercise, identifying the part that has not been done correctly, and a real-time performance correction. In this way, the recommender system receives a greater amount of information and so can more accurately indicate the exercise to be performed. In terms of the recommender system, an algorithm was developed to weigh the performance, temporal evolution and preferences, providing a more accurate recommendation, as well as expanding the recommendation to a batch of exercises, instead of just one., This work was partly supported by the FCT-Fundacao para a Ciencia e Tecnologia through the Post-Doc scholarship SFRH/BPD/102696/2014 and by the Spanish Government TIN2016-76515-R Grant supported with Feder funds.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

PHAROS-PHysical Assistant RObot System

  • Costa, Angelo
  • Martinez-Martin, Ester
  • Cazorla, M.
  • Julian, Vicente|||0000-0002-2743-6037
[EN] The great demographic change leading to an ageing society demands technological solutions to satisfy the increasing varied elderly needs. This paper presents PHAROS, an interactive robot system that recommends and monitors physical exercises designed for the elderly. The aim of PHAROS is to be a friendly elderly companion that periodically suggests personalised physical activities, promoting healthy living and active ageing. Here, it is presented the PHAROS architecture, components and experimental results. The architecture has three main strands: a Pepper robot, that interacts with the users and records their exercises performance; the Human Exercise Recognition, that uses the Pepper recorded information to classify the exercise performed using Deep Leaning methods; and the Recommender, a smart-decision maker that schedules periodically personalised physical exercises in the users' agenda. The experimental results show a high accuracy in terms of detecting and classifying the physical exercises (97.35%) done by 7 persons. Furthermore, we have implemented a novel procedure of rating exercises on the recommendation algorithm. It closely follows the users' health status (poor performance may reveal health problems) and adapts the suggestions to it. The history may be used to access the physical condition of the user, revealing underlying problems that may be impossible to see otherwise., This work was partly supported by the FCT-Fundacao para a Ciencia e Tecnologia through the Post-Doc scholarship SFRH/BPD/102696/2014 and by the Spanish Government TIN2016-76515-R and TIN2015-65515-C4-1-R Grants supported with Feder funds.
RUA. Repositorio Institucional de la Universidad de Alicante
Publication Otros (Other)

Semi-supervised 3D object recognition through CNN labeling

  • Rangel, José Carlos
  • Martínez-Gómez, Jesús
  • Romero-González, Cristina
  • García-Varea, Ismael
  • Cazorla, Miguel
  • Cazorla, Miguel|||/items/bd003579-c106-4ca8-814b-000fe9d1b9a1
Despite the outstanding results of Convolutional Neural Networks (CNNs) in object recognition and classification, there are still some open problems to address when applying these solutions to real-world problems. Specifically, CNNs struggle to generalize under challenging scenarios, like recognizing the variability and heterogeneity of the instances of elements belonging to the same category. Some of these difficulties are directly related to the input information, 2D-based methods still show a lack of robustness against strong lighting variations, for example. In this paper, we propose to merge techniques using both 2D and 3D information to overcome these problems. Specifically, we take advantage of the spatial information in the 3D data to segment objects in the image and build an object classifier, and the classification capabilities of CNNs to semi-supervisedly label each object image for training. As the experimental results demonstrate, our model can successfully generalize for categories with high intra-class variability and outperform the accuracy of a well-known CNN model.
RUA. Repositorio Institucional de la Universidad de Alicante
Publication Otros (Other)

Inferring Static Hand Poses from a Low-Cost Non-Intrusive sEMG Sensor

  • Nasri, Nadia
  • Orts-Escolano, Sergio
  • Gomez-Donoso, Francisco
  • Cazorla, Miguel
  • Cazorla, Miguel|||/items/bd003579-c106-4ca8-814b-000fe9d1b9a1
  • Gomez-Donoso, Francisco|||/items/13200984-24ea-4064-a994-a6defdc16d58
  • Orts-Escolano, Sergio|||/items/2222768a-df75-443a-89ea-0ca8ede0e69b
Every year, a significant number of people lose a body part in an accident, through sickness or in high-risk manual jobs. Several studies and research works have tried to reduce the constraints and risks in their lives through the use of technology. This work proposes a learning-based approach that performs gesture recognition using a surface electromyography-based device, the Myo Armband released by Thalmic Labs, which is a commercial device and has eight non-intrusive low-cost sensors. With 35 able-bodied subjects, and using the Myo Armband device, which is able to record data at about 200 MHz, we collected a dataset that includes six dissimilar hand gestures. We used a gated recurrent unit network to train a system that, as input, takes raw signals extracted from the surface electromyography sensors. The proposed approach obtained a 99.90% training accuracy and 99.75% validation accuracy. We also evaluated the proposed system on a test set (new subjects) obtaining an accuracy of 77.85%. In addition, we showed the test prediction results for each gesture separately and analyzed which gestures for the Myo armband with our suggested network can be difficult to distinguish accurately. Moreover, we studied for first time the gated recurrent unit network capability in gesture recognition approaches. Finally, we integrated our method in a system that is able to classify live hand gestures.
RUA. Repositorio Institucional de la Universidad de Alicante
Publication Otros (Other)

PHAROS—PHysical Assistant RObot System

  • Costa, Angelo
  • Martinez-Martin, Ester
  • Cazorla, Miguel
  • Julián, Vicente
  • Cazorla, Miguel|||/items/bd003579-c106-4ca8-814b-000fe9d1b9a1
  • Costa, Angelo|||/items/f4ecad24-bcc1-47f6-a263-6b525558ae6c
  • Martinez-Martin, Ester|||/items/74a8fa4f-d7ad-4df1-9b91-355b474a0cd2
The great demographic change leading to an ageing society demands technological solutions to satisfy the increasing varied elderly needs. This paper presents PHAROS, an interactive robot system that recommends and monitors physical exercises designed for the elderly. The aim of PHAROS is to be a friendly elderly companion that periodically suggests personalised physical activities, promoting healthy living and active ageing. Here, it is presented the PHAROS architecture, components and experimental results. The architecture has three main strands: a Pepper robot, that interacts with the users and records their exercises performance; the Human Exercise Recognition, that uses the Pepper recorded information to classify the exercise performed using Deep Leaning methods; and the Recommender, a smart-decision maker that schedules periodically personalised physical exercises in the users’ agenda. The experimental results show a high accuracy in terms of detecting and classifying the physical exercises (97.35%) done by 7 persons. Furthermore, we have implemented a novel procedure of rating exercises on the recommendation algorithm. It closely follows the users’ health status (poor performance may reveal health problems) and adapts the suggestions to it. The history may be used to access the physical condition of the user, revealing underlying problems that may be impossible to see otherwise.
RUA. Repositorio Institucional de la Universidad de Alicante
Publication Otros (Other)

Complex Methods Applied to Data Analysis, Processing, and Visualisation

  • Garcia-Rodriguez, Jose
  • Angelopoulou, Anastassia
  • Tomás, David
  • Lewis, Andrew
  • Garcia-Rodriguez, Jose|||/items/d5bc18f8-44db-4723-b291-5849321e9e69
  • Tomás, David|||/items/cf031fc3-3c30-454b-a6df-3762c29cc296
RUA. Repositorio Institucional de la Universidad de Alicante
Publication Otros (Other)

Pedestrian Movement Direction Recognition Using Convolutional Neural Networks

  • Dominguez-Sanchez, Alex
  • Cazorla, Miguel
  • Orts-Escolano, Sergio
  • Cazorla, Miguel|||/items/bd003579-c106-4ca8-814b-000fe9d1b9a1
  • Orts-Escolano, Sergio|||/items/2222768a-df75-443a-89ea-0ca8ede0e69b
Pedestrian movement direction recognition is an important factor in autonomous driver assistance and security surveillance systems. Pedestrians are the most crucial and fragile moving objects in streets, roads, and events, where thousands of people may gather on a regular basis. People flow analysis on zebra crossings and in shopping centers or events such as demonstrations are a key element to improve safety and to enable autonomous cars to drive in real life environments. This paper focuses on deep learning techniques such as convolutional neural networks (CNN) to achieve a reliable detection of pedestrians moving in a particular direction. We propose a CNN-based technique that leverages current pedestrian detection techniques (histograms of oriented gradients-linSVM) to generate a sum of subtracted frames (flow estimation around the detected pedestrian), which are used as an input for the proposed modified versions of various state-of-the-art CNN networks, such as AlexNet, GoogleNet, and ResNet. Moreover, we have also created a new data set for this purpose, and analyzed the importance of training in a known data set for the neural networks to achieve reliable results.
RUA. Repositorio Institucional de la Universidad de Alicante
Publication Otros (Other)

Large-scale Multiview 3D Hand Pose Dataset

  • Gomez-Donoso, Francisco
  • Orts-Escolano, Sergio
  • Cazorla, Miguel
  • Cazorla, Miguel|||/items/bd003579-c106-4ca8-814b-000fe9d1b9a1
  • Gomez-Donoso, Francisco|||/items/13200984-24ea-4064-a994-a6defdc16d58
  • Orts-Escolano, Sergio|||/items/2222768a-df75-443a-89ea-0ca8ede0e69b
Accurate visual hand pose estimation at joint level has several applications for human-robot interaction, natural user interfaces and virtual/augmented reality applications. However, it is still an open problem being addressed by the computer vision community. Recent novel deep learning techniques may help circumvent the limitations of standard approaches. However, they require large amounts of accurate annotated data. Hand pose datasets that have been released so far present issues such as limited number of samples, inaccurate data or high-level annotations. Moreover, most of them are focused on depth-based approaches, providing only depth information (missing RGB data). In this work, we present a novel multiview hand pose dataset in which we provide hand color images and different kind of annotations for each sample, i.e. the bounding box and the 2D and 3D location on the joints in the hand. Furthermore, we introduce a simple yet accurate deep learning architecture for real-time robust 2D hand pose estimation. Then, we conduct experiments that show how the use of the proposed dataset in the training stage produces accurate results for 2D hand pose estimation using a single color camera.
RUA. Repositorio Institucional de la Universidad de Alicante
Publication Otros (Other)

PHAROS 2.0—A PHysical Assistant RObot System Improved

  • Martinez-Martin, Ester
  • Costa, Angelo
  • Cazorla, Miguel
  • Cazorla, Miguel|||/items/bd003579-c106-4ca8-814b-000fe9d1b9a1
  • Costa, Angelo|||/items/f4ecad24-bcc1-47f6-a263-6b525558ae6c
  • Martinez-Martin, Ester|||/items/74a8fa4f-d7ad-4df1-9b91-355b474a0cd2
There are great physical and cognitive benefits for older adults who are engaged in active aging, a process that should involve daily exercise. In our previous work on the PHysical Assistant RObot System (PHAROS), we developed a system that proposed and monitored physical activities. The system used a social robot to analyse, by means of computer vision, the exercise a person was doing. Then, a recommender system analysed the exercise performed and indicated what exercise to perform next. However, the system needed certain improvements. On the one hand, the vision system captured the movement of the person and indicated whether the exercise had been done correctly or not. On the other hand, the recommender system was based purely on a ranking system that did not take into account temporal evolution and preferences. In this work, we propose an evolution of PHAROS, PHAROS 2.0, incorporating improvements in both of the previously mentioned aspects. In the motion capture aspect, we are now able to indicate the degree of completeness of each exercise, identifying the part that has not been done correctly, and a real-time performance correction. In this way, the recommender system receives a greater amount of information and so can more accurately indicate the exercise to be performed. In terms of the recommender system, an algorithm was developed to weigh the performance, temporal evolution and preferences, providing a more accurate recommendation, as well as expanding the recommendation to a batch of exercises, instead of just one.
RUA. Repositorio Institucional de la Universidad de Alicante
Publication Otros (Other)

A Socially Assistive Robot for Elderly Exercise Promotion

  • Martinez-Martin, Ester
  • Cazorla, Miguel
  • Cazorla, Miguel|||/items/bd003579-c106-4ca8-814b-000fe9d1b9a1
  • Martinez-Martin, Ester|||/items/74a8fa4f-d7ad-4df1-9b91-355b474a0cd2
The population ageing phenomenon leads to an unceasing need for home-based healthcare systems to continuously monitor the elderly’s cognitive and physical health. In this sense, physical activity may be beneficial in preserving cognition in elder life as well as in providing clinicians and therapists with the indicative of elderly’s health condition. Nevertheless, current systems fail to promote and monitor the elderly’s physical activity in their living environments. This paper is aimed at providing a socially assistive robot solution for this task. Since robot acceptance depends to a great extent on its robustness in performing tasks, we have focused on exercise recognition due to its great importance for both clinicians and elderly. For that, two different tasks were carried out. First, an image dataset for physical exercise recognition has been generated. Then, a comparative analysis of several deep learning techniques is presented. This paper reveals a great performance in the exercise recognition of CNN-LSTM with an exercise recognition accuracy of 99.87%.
RUA. Repositorio Institucional de la Universidad de Alicante
Publication Otros (Other)

Compression and registration of 3D point clouds using GMMs

  • Navarrete, Javier
  • Viejo Hernando, Diego
  • Cazorla, Miguel
  • Cazorla, Miguel|||/items/bd003579-c106-4ca8-814b-000fe9d1b9a1
  • Viejo Hernando, Diego|||/items/2c1f802c-1c6e-4aef-9911-82a878de0a39
3D data sensors provide an enormous amount of information. It is necessary to develop efficient methods to manage this information under certain time, bandwidth or storage space requirements. In this work, we propose a 3D compression and decompression method. This method also allows the use of the compressed data for a registration process. First, points are selected and grouped, using a 3D-model based on planar surfaces. Next, we use a fast variant of Gaussian Mixture Models and an Expectation-Maximization algorithm to replace the points grouped in the previous step with a set of Gaussian distributions. These learned models can be used as features to find matches between two consecutive poses and apply 3D pose registration using RANSAC. Finally, the 3D map can be obtained by decompressing the models.
RUA. Repositorio Institucional de la Universidad de Alicante
Publication Otros (Other)

Automatic Learning Improves Human-Robot Interaction in Productive Environments: A Review

  • Zamora Hernández, Mauricio Andrés
  • Caldwell Marín, Eldon Glen
  • Garcia-Rodriguez, Jose
  • Azorin-Lopez, Jorge
  • Cazorla, Miguel
  • Azorin-Lopez, Jorge|||/items/d43d029b-14bf-4dac-9bda-ed08e213bb1d
  • Cazorla, Miguel|||/items/bd003579-c106-4ca8-814b-000fe9d1b9a1
  • Garcia-Rodriguez, Jose|||/items/d5bc18f8-44db-4723-b291-5849321e9e69
In the creation of new industries, products and services -- all of which are advances of the Fourth Industrial Revolution -- the human-robot interaction that includes automatic learning and computer vision are elements to consider since they promote collaborative environments between people and robots. The use of machine learning and computer vision provides the tools needed to increase productivity and minimizes delivery reaction times by assisting in the optimization of complex production planning processes. This review of the state of the art presents the main trends that seek to improve human-robot interaction in productive environments, and identifies challenges in research as well as in industrial - technological development in this topic. In addition, this review offers a proposal on the needs of use of artificial intelligence in all processes of industry 4.0 as a crucial linking element among humans, robots, intelligent and traditional machines; as well as a mechanism for quality control and occupational safety.
RUA. Repositorio Institucional de la Universidad de Alicante
Publication Otros (Other)

Classifying Behaviours in Videos with Recurrent Neural Networks

  • Abellan-Abenza, Javier
  • Garcia-Garcia, Alberto
  • Oprea, Sergiu
  • Ivorra-Piqueres, David
  • Garcia-Rodriguez, Jose
  • Garcia-Rodriguez, Jose|||/items/d5bc18f8-44db-4723-b291-5849321e9e69
This article describes how the human activity recognition in videos is a very attractive topic among researchers due to vast possible applications. This article considers the analysis of behaviors and activities in videos obtained with low-cost RGB cameras. To do this, a system is developed where a video is input, and produces as output the possible activities happening in the video. This information could be used in many applications such as video surveillance, disabled person assistance, as a home assistant, employee monitoring, etc. The developed system makes use of the successful techniques of Deep Learning. In particular, convolutional neural networks are used to detect features in the video images, meanwhile Recurrent Neural Networks are used to analyze these features and predict the possible activity in the video.
RUA. Repositorio Institucional de la Universidad de Alicante
Publication Otros (Other)

Rehabilitation Technology: Assistance from Hospital to Home

  • Martinez-Martin, Ester
  • Cazorla, Miguel
  • Cazorla, Miguel|||/items/bd003579-c106-4ca8-814b-000fe9d1b9a1
  • Martinez-Martin, Ester|||/items/74a8fa4f-d7ad-4df1-9b91-355b474a0cd2
Rehabilitation is essential for disabled people to achieve the highest level of functional independence, reducing or preventing impairments. Nonetheless, this process can be long and expensive. This fact together with the ageing phenomenon has become a critical issue for both clinicians and patients. In this sense, technological solutions may be beneficial since they reduce the costs and increase the number of patients per caregiver, which makes them more accessible. In addition, they provide access to rehabilitation services for those facing physical, financial, and/or attitudinal barriers. This paper presents the state of the art of the assistive rehabilitation technologies for different recovery methods starting from in-person sessions to complementary at-home activities.
RUA. Repositorio Institucional de la Universidad de Alicante
Publication Otros (Other)

Enhancing perception for the visually impaired with deep learning techniques and low-cost wearable sensors

  • Bauer, Zuria
  • Dominguez, Alejandro
  • Cruz, Edmanuel
  • Gomez-Donoso, Francisco
  • Orts-Escolano, Sergio
  • Cazorla, Miguel
  • Cazorla, Miguel|||/items/bd003579-c106-4ca8-814b-000fe9d1b9a1
  • Gomez-Donoso, Francisco|||/items/13200984-24ea-4064-a994-a6defdc16d58
  • Orts-Escolano, Sergio|||/items/2222768a-df75-443a-89ea-0ca8ede0e69b
As estimated by the World Health Organization, there are millions of people who lives with some form of vision impairment. As a consequence, some of them present mobility problems in outdoor environments. With the aim of helping them, we propose in this work a system which is capable of delivering the position of potential obstacles in outdoor scenarios. Our approach is based on non-intrusive wearable devices and focuses also on being low-cost. First, a depth map of the scene is estimated from a color image, which provides 3D information of the environment. Then, an urban object detector is in charge of detecting the semantics of the objects in the scene. Finally, the three-dimensional and semantic data is summarized in a simpler representation of the potential obstacles the users have in front of them. This information is transmitted to the user through spoken or haptic feedback. Our system is able to run at about 3.8 fps and achieved a 87.99% mean accuracy in obstacle presence detection. Finally, we deployed our system in a pilot test which involved an actual person with vision impairment, who validated the effectiveness of our proposal for improving its navigation capabilities in outdoors.
RUA. Repositorio Institucional de la Universidad de Alicante
Publication Otros (Other)

A New Dataset and Performance Evaluation of a Region-Based CNN for Urban Object Detection

  • Dominguez-Sanchez, Alex
  • Cazorla, Miguel
  • Orts-Escolano, Sergio
  • Cazorla, Miguel|||/items/bd003579-c106-4ca8-814b-000fe9d1b9a1
  • Orts-Escolano, Sergio|||/items/2222768a-df75-443a-89ea-0ca8ede0e69b
In recent years, we have seen a large growth in the number of applications which use deep learning-based object detectors. Autonomous driving assistance systems (ADAS) are one of the areas where they have the most impact. This work presents a novel study evaluating a state-of-the-art technique for urban object detection and localization. In particular, we investigated the performance of the Faster R-CNN method to detect and localize urban objects in a variety of outdoor urban videos involving pedestrians, cars, bicycles and other objects moving in the scene (urban driving). We propose a new dataset that is used for benchmarking the accuracy of a real-time object detector (Faster R-CNN). Part of the data was collected using an HD camera mounted on a vehicle. Furthermore, some of the data is weakly annotated so it can be used for testing weakly supervised learning techniques. There already exist urban object datasets, but none of them include all the essential urban objects. We carried out extensive experiments demonstrating the effectiveness of the baseline approach. Additionally, we propose an R-CNN plus tracking technique to accelerate the process of real-time urban object detection.