2021
|
Borja-Borja, Luis Felipe; Azorín-López, Jorge; Saval-Calvo, Marcelo A Deep Learning Architecture for Recognizing Abnormal Activities of Groups Using Context and Motion Information Artículo en actas En: 15th International Conference on Soft Computing Models in Industrial and Environmental Applications, pp. 760–769, 2021. @inproceedings{Borja-Borja2021,
title = {A Deep Learning Architecture for Recognizing Abnormal Activities of Groups Using Context and Motion Information},
author = {Luis Felipe Borja-Borja and Jorge Azorín-López and Marcelo Saval-Calvo},
url = {http://link.springer.com/10.1007/978-3-030-57802-2_73},
doi = {10.1007/978-3-030-57802-2_73},
year = {2021},
date = {2021-01-01},
urldate = {2021-01-01},
booktitle = {15th International Conference on Soft Computing Models in Industrial and Environmental Applications},
pages = {760--769},
abstract = {Currently, the automation of activity recognition of a group of people in closed and open environments is a major problem, especially in video surveillance. It is becoming increasingly important to have computer vision architectures that allow automatic recognition of group activities to make decisions. This paper proposes a computer vision architecture capable of learning and recognizing abnormal group activities using the movements of the group in the scene. It is based on the Activity Description Vector, a descriptor capable of representing the trajectory information of a sequence of images as a collection of local movements that occur in specific regions of the scene. The proposal is based on the evolution of different versions of this descriptor towards the generation of images that will be input of a two-stream classifier capable of robustly classifying abnormal group activities. Moreover, it includes context information to provide extra information to classify the activities including it as the third stream of the classifier resulting in a robust architecture for one class classification problems. The architecture has been evaluated and compared with other approaches using Ped 1 and Ped 2 datasets, obtaining a high performance in abnormal group activity recognition.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Currently, the automation of activity recognition of a group of people in closed and open environments is a major problem, especially in video surveillance. It is becoming increasingly important to have computer vision architectures that allow automatic recognition of group activities to make decisions. This paper proposes a computer vision architecture capable of learning and recognizing abnormal group activities using the movements of the group in the scene. It is based on the Activity Description Vector, a descriptor capable of representing the trajectory information of a sequence of images as a collection of local movements that occur in specific regions of the scene. The proposal is based on the evolution of different versions of this descriptor towards the generation of images that will be input of a two-stream classifier capable of robustly classifying abnormal group activities. Moreover, it includes context information to provide extra information to classify the activities including it as the third stream of the classifier resulting in a robust architecture for one class classification problems. The architecture has been evaluated and compared with other approaches using Ped 1 and Ped 2 datasets, obtaining a high performance in abnormal group activity recognition. |
2020
|
Villena-Martinez, Victor; Oprea, Sergiu; Saval-Calvo, Marcelo; Azorin-Lopez, Jorge; Fuster-Guillo, Andres; Fisher, Robert B When Deep Learning Meets Data Alignment: A Review on Deep Registration Networks (DRNs) Artículo de revista En: Applied Sciences, vol. 10, no. 21, pp. 7524, 2020, ISSN: 2076-3417. @article{Villena-Martinez2020,
title = {When Deep Learning Meets Data Alignment: A Review on Deep Registration Networks (DRNs)},
author = {Victor Villena-Martinez and Sergiu Oprea and Marcelo Saval-Calvo and Jorge Azorin-Lopez and Andres Fuster-Guillo and Robert B Fisher},
url = {https://www.mdpi.com/2076-3417/10/21/7524},
doi = {10.3390/app10217524},
issn = {2076-3417},
year = {2020},
date = {2020-10-01},
journal = {Applied Sciences},
volume = {10},
number = {21},
pages = {7524},
abstract = {This paper reviews recent deep learning-based registration methods. Registration is the process that computes the transformation that aligns datasets, and the accuracy of the result depends on multiple factors. The most significant factors are the size of input data; the presence of noise, outliers and occlusions; the quality of the extracted features; real-time requirements; and the type of transformation, especially those defined by multiple parameters, such as non-rigid deformations. Deep Registration Networks (DRNs) are those architectures trying to solve the alignment task using a learning algorithm. In this review, we classify these methods according to a proposed framework based on the traditional registration pipeline. This pipeline consists of four steps: target selection, feature extraction, feature matching, and transform computation for the alignment. This new paradigm introduces a higher-level understanding of registration, which makes explicit the challenging problems of traditional approaches. The main contribution of this work is to provide a comprehensive starting point to address registration problems from a learning-based perspective and to understand the new range of possibilities.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
This paper reviews recent deep learning-based registration methods. Registration is the process that computes the transformation that aligns datasets, and the accuracy of the result depends on multiple factors. The most significant factors are the size of input data; the presence of noise, outliers and occlusions; the quality of the extracted features; real-time requirements; and the type of transformation, especially those defined by multiple parameters, such as non-rigid deformations. Deep Registration Networks (DRNs) are those architectures trying to solve the alignment task using a learning algorithm. In this review, we classify these methods according to a proposed framework based on the traditional registration pipeline. This pipeline consists of four steps: target selection, feature extraction, feature matching, and transform computation for the alignment. This new paradigm introduces a higher-level understanding of registration, which makes explicit the challenging problems of traditional approaches. The main contribution of this work is to provide a comprehensive starting point to address registration problems from a learning-based perspective and to understand the new range of possibilities. |
Fuster-Guilló, Andrés; Azorín-López, Jorge; Saval-Calvo, Marcelo; Castillo-Zaragoza, Juan Miguel; Garcia-D'Urso, Nahuel; Fisher, Robert B RGB-D-Based Framework to Acquire, Visualize and Measure the Human Body for Dietetic Treatments Artículo de revista En: Sensors, vol. 20, no. 13, pp. 3690, 2020, ISSN: 1424-8220. @article{Fuster-Guillo2020,
title = {RGB-D-Based Framework to Acquire, Visualize and Measure the Human Body for Dietetic Treatments},
author = {Andrés Fuster-Guilló and Jorge Azorín-López and Marcelo Saval-Calvo and Juan Miguel Castillo-Zaragoza and Nahuel Garcia-D'Urso and Robert B Fisher},
url = {https://www.mdpi.com/1424-8220/20/13/3690},
doi = {10.3390/s20133690},
issn = {1424-8220},
year = {2020},
date = {2020-07-01},
journal = {Sensors},
volume = {20},
number = {13},
pages = {3690},
abstract = {This research aims to improve dietetic-nutritional treatment using state-of-the-art RGB-D sensors and virtual reality (VR) technology. Recent studies show that adherence to treatment can be improved using multimedia technologies. However, there are few studies using 3D data and VR technologies for this purpose. On the other hand, obtaining 3D measurements of the human body and analyzing them over time (4D) in patients undergoing dietary treatment is a challenging field. The main contribution of the work is to provide a framework to study the effect of 4D body model visualization on adherence to obesity treatment. The system can obtain a complete 3D model of a body using low-cost technology, allowing future straightforward transference with sufficient accuracy and realistic visualization, enabling the analysis of the evolution (4D) of the shape during the treatment of obesity. The 3D body models will be used for studying the effect of visualization on adherence to obesity treatment using 2D and VR devices. Moreover, we will use the acquired 3D models to obtain measurements of the body. An analysis of the accuracy of the proposed methods for obtaining measurements with both synthetic and real objects has been carried out.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
This research aims to improve dietetic-nutritional treatment using state-of-the-art RGB-D sensors and virtual reality (VR) technology. Recent studies show that adherence to treatment can be improved using multimedia technologies. However, there are few studies using 3D data and VR technologies for this purpose. On the other hand, obtaining 3D measurements of the human body and analyzing them over time (4D) in patients undergoing dietary treatment is a challenging field. The main contribution of the work is to provide a framework to study the effect of 4D body model visualization on adherence to obesity treatment. The system can obtain a complete 3D model of a body using low-cost technology, allowing future straightforward transference with sufficient accuracy and realistic visualization, enabling the analysis of the evolution (4D) of the shape during the treatment of obesity. The 3D body models will be used for studying the effect of visualization on adherence to obesity treatment using 2D and VR devices. Moreover, we will use the acquired 3D models to obtain measurements of the body. An analysis of the accuracy of the proposed methods for obtaining measurements with both synthetic and real objects has been carried out. |
Borja-Borja, Luis Felipe; Azorin-Lopez, Jorge; Saval-Calvo, Marcelo; Fuster-Guillo, Andres Deep Learning Architecture for Group Activity Recognition using Description of Local Motions Artículo en actas En: 2020 International Joint Conference on Neural Networks (IJCNN), pp. 1–8, IEEE, 2020, ISBN: 978-1-7281-6926-2. @inproceedings{Borja-Borja2020,
title = {Deep Learning Architecture for Group Activity Recognition using Description of Local Motions},
author = {Luis Felipe Borja-Borja and Jorge Azorin-Lopez and Marcelo Saval-Calvo and Andres Fuster-Guillo},
url = {https://ieeexplore.ieee.org/document/9207366/},
doi = {10.1109/IJCNN48605.2020.9207366},
isbn = {978-1-7281-6926-2},
year = {2020},
date = {2020-07-01},
booktitle = {2020 International Joint Conference on Neural Networks (IJCNN)},
pages = {1--8},
publisher = {IEEE},
abstract = {Nowadays, the recognition of group activities is a significant problem, specially in video surveillance. It is increasingly important to have vision architectures that automatically allow timely recognition of group activities and predictions about them in order to make decisions. This paper proposes a computer vision architecture able to learn and recognise group activities using the movements of it in the scene. It is based on the Activity Description Vector (ADV), a descriptor able to represent the trajectory information of an image sequence as a collection of the local movements that occur in specific regions of the scene. The proposal evolves this descriptor towards the generation of images able to be the input queue of a two-stream convolutional neural network capable of robustly classifying group activities. Hence, this proposal, besides the use of trajectory analysis that allows a simple high level understanding of complex groups activities, takes advantage of the deep learning characteristics providing a robust architecture for multi-class recognition. The architecture has been evaluated and compared to other approaches using BEHAVE and INRIA dataset sequences obtaining great performance in the recognition of group activities.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Nowadays, the recognition of group activities is a significant problem, specially in video surveillance. It is increasingly important to have vision architectures that automatically allow timely recognition of group activities and predictions about them in order to make decisions. This paper proposes a computer vision architecture able to learn and recognise group activities using the movements of it in the scene. It is based on the Activity Description Vector (ADV), a descriptor able to represent the trajectory information of an image sequence as a collection of the local movements that occur in specific regions of the scene. The proposal evolves this descriptor towards the generation of images able to be the input queue of a two-stream convolutional neural network capable of robustly classifying group activities. Hence, this proposal, besides the use of trajectory analysis that allows a simple high level understanding of complex groups activities, takes advantage of the deep learning characteristics providing a robust architecture for multi-class recognition. The architecture has been evaluated and compared to other approaches using BEHAVE and INRIA dataset sequences obtaining great performance in the recognition of group activities. |
Mora, Higinio; Sirvent, Rafael Alejandro Mollá; Azorin-Lopez, Jorge; Fuster-Guilló, Andrés; Sanchez-Romero, Jose-Luis; Pujol, Francisco A |; Garcia-Rodriguez, Jose; Jimeno-Morenilla, Antonio; Saval-Calvo, Marcelo; Garcia-Garcia, Alberto; Martínez, Víctor Villena Impacto de los Sistemas de Información Geográfica en la motivación y emprendimiento en la materia de Internet de las Cosas Artículo de revista En: La docencia en la Enseñanza Superior. Nuevas aportaciones desde la investigación e innovación educativas., pp. 297–307, 2020. @article{Mora2020,
title = {Impacto de los Sistemas de Información Geográfica en la motivación y emprendimiento en la materia de Internet de las Cosas},
author = {Higinio Mora and Rafael Alejandro Mollá Sirvent and Jorge Azorin-Lopez and Andrés Fuster-Guilló and Jose-Luis Sanchez-Romero and Francisco A | Pujol and Jose Garcia-Rodriguez and Antonio Jimeno-Morenilla and Marcelo Saval-Calvo and Alberto Garcia-Garcia and Víctor Villena Martínez},
year = {2020},
date = {2020-01-01},
journal = {La docencia en la Enseñanza Superior. Nuevas aportaciones desde la investigación e innovación educativas.},
pages = {297--307},
abstract = {Las nuevas tecnologías de localización y geoposicionamiento forman parte del conjunto de tecnologías disruptivas que están transformando la sociedad actual. Aprender a utilizar estas tecnologías e introducirlas en las aulas proporciona ventajas significativas en el proceso de enseñanza-aprendizaje y tiene beneficios potenciales para los estudiantes, ya que muchas de estas tecnologías son esenciales en numerosas profesiones actuales, además, abre posibilidades para el desarrollo de proyectos en línea con los nuevos avances tecnológicos. Sin embargo, la utilización de los Sistemas de Información Geográfica y los sistemas de localización presentan problemas para su correcta aplicación en la docencia (falta de homogenización, información incompleta, etc.). Por consiguiente, en este proyecto analizamos los Sistemas de Información Geográfica y las distintas tecnologías de localización y estudiamos su potencial en la adquisición de conocimientos y habilidades para la resolución de problemas de ingeniería informática ampliando la visión de negocio y de aplicabilidad de los alumnos. El método utilizado se basa fundamentalmente en el análisis cualitativo realizado por el equipo docente de las asignaturas basado en su experiencia y en el análisis fenomenológico. Si bien las calificaciones de los estudiantes son muy similares a cursos anteriores, la calidad de los trabajos presentados durante este curso es superior. Adicionalmente, se constata una mayor motivación, más implicación de los estudiantes y una creación de visión de futuro en cuanto a su desarrollo profesional.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Las nuevas tecnologías de localización y geoposicionamiento forman parte del conjunto de tecnologías disruptivas que están transformando la sociedad actual. Aprender a utilizar estas tecnologías e introducirlas en las aulas proporciona ventajas significativas en el proceso de enseñanza-aprendizaje y tiene beneficios potenciales para los estudiantes, ya que muchas de estas tecnologías son esenciales en numerosas profesiones actuales, además, abre posibilidades para el desarrollo de proyectos en línea con los nuevos avances tecnológicos. Sin embargo, la utilización de los Sistemas de Información Geográfica y los sistemas de localización presentan problemas para su correcta aplicación en la docencia (falta de homogenización, información incompleta, etc.). Por consiguiente, en este proyecto analizamos los Sistemas de Información Geográfica y las distintas tecnologías de localización y estudiamos su potencial en la adquisición de conocimientos y habilidades para la resolución de problemas de ingeniería informática ampliando la visión de negocio y de aplicabilidad de los alumnos. El método utilizado se basa fundamentalmente en el análisis cualitativo realizado por el equipo docente de las asignaturas basado en su experiencia y en el análisis fenomenológico. Si bien las calificaciones de los estudiantes son muy similares a cursos anteriores, la calidad de los trabajos presentados durante este curso es superior. Adicionalmente, se constata una mayor motivación, más implicación de los estudiantes y una creación de visión de futuro en cuanto a su desarrollo profesional. |
2019
|
Fuster-Guilló, Andrés; Azorín-López, Jorge; Zaragoza, Juan Miguel Castillo; Pérez, Luis Fernando Pérez; Saval-Calvo, Marcelo; Fisher, Robert B 3D Technologies to Acquire and Visualize the Human Body for Improving Dietetic Treatment Artículo en actas En: 13th International Conference on Ubiquitous Computing and Ambient Intelligence UCAmI, pp. 53, 2019, ISSN: 2504-3900. @inproceedings{Fuster-Guillo2019,
title = {3D Technologies to Acquire and Visualize the Human Body for Improving Dietetic Treatment},
author = {Andrés Fuster-Guilló and Jorge Azorín-López and Juan Miguel Castillo Zaragoza and Luis Fernando Pérez Pérez and Marcelo Saval-Calvo and Robert B Fisher},
url = {https://www.mdpi.com/2504-3900/31/1/53},
doi = {10.3390/proceedings2019031053},
issn = {2504-3900},
year = {2019},
date = {2019-11-01},
booktitle = {13th International Conference on Ubiquitous Computing and Ambient Intelligence UCAmI},
volume = {31},
number = {1},
pages = {53},
abstract = {This research aims to improve adherence to dietetic-nutritional treatment using state-of-the-art RGB-D sensor and virtual reality (VR) technology. Recent studies show that adherence to treatment can be improved by using multimedia technologies which impact on the body awareness of patients. However, there are no studies published to date using 3D data and VR technologies for this purpose. This paper describes a system capable of obtaining the complete 3D model of a body with high accuracy and a realistic visualization for 2D and VR devices to be used for studying the effect of 3D technologies on adherence to obesity treatment.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
This research aims to improve adherence to dietetic-nutritional treatment using state-of-the-art RGB-D sensor and virtual reality (VR) technology. Recent studies show that adherence to treatment can be improved by using multimedia technologies which impact on the body awareness of patients. However, there are no studies published to date using 3D data and VR technologies for this purpose. This paper describes a system capable of obtaining the complete 3D model of a body with high accuracy and a realistic visualization for 2D and VR devices to be used for studying the effect of 3D technologies on adherence to obesity treatment. |
Mora, Higinio; Mollá-Sirvent, Rafael; Azorín-López, Jorge; Fuster-Guilló, Andrés; Jimeno-Morenilla, Anotnio Manuel; Sánchez-Romero, José Luis; Pujol-López, Francisco Antonio; García-Rodríguez, José; Saval-Calvo, Marcelo; Villena-Martínez, Victor; García-García, Alberto NEW WEB 3.0 TREND FOR THE IMPROVEMENT OF TEACHING-LEARNING PROCESSES Artículo en actas En: 13th International Technology, Education and Development Conference, pp. 2990–2995, 2019. @inproceedings{Mora2019,
title = {NEW WEB 3.0 TREND FOR THE IMPROVEMENT OF TEACHING-LEARNING PROCESSES},
author = {Higinio Mora and Rafael Mollá-Sirvent and Jorge Azorín-López and Andrés Fuster-Guilló and Anotnio Manuel Jimeno-Morenilla and José Luis Sánchez-Romero and Francisco Antonio Pujol-López and José García-Rodríguez and Marcelo Saval-Calvo and Victor Villena-Martínez and Alberto García-García},
url = {http://library.iated.org/view/MORA2019NEW},
doi = {10.21125/inted.2019.0789},
year = {2019},
date = {2019-03-01},
urldate = {2019-03-01},
booktitle = {13th International Technology, Education and Development Conference},
pages = {2990--2995},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
|
Saval-Calvo, Marcelo; Villena-Martínez, Victor; Azorín-López, Jorge; Fuster-Guilló, Andrés; García-García, Alberto; Jimeno-Morenilla, Antonio Manuel; García-Rodríguez, José; Pujol-López, Francisco Antonio; Mora-Mora, Higinio; Sánchez-Romero, Jose Luís; Mollá-Sirvent, Rafael STUDY OF THE PERFORMANCE EVOLUTION OVER TIME IN HIGH ACADEMIC PERFORMANCE GROUPS IN COMPUTER ARCHITECTURE COURSE Artículo en actas En: 13th International Technology, Education and Development Conference, pp. 2981–2988, 2019. @inproceedings{Saval-Calvo2019,
title = {STUDY OF THE PERFORMANCE EVOLUTION OVER TIME IN HIGH ACADEMIC PERFORMANCE GROUPS IN COMPUTER ARCHITECTURE COURSE},
author = {Marcelo Saval-Calvo and Victor Villena-Martínez and Jorge Azorín-López and Andrés Fuster-Guilló and Alberto García-García and Antonio Manuel Jimeno-Morenilla and José García-Rodríguez and Francisco Antonio Pujol-López and Higinio Mora-Mora and Jose Luís Sánchez-Romero and Rafael Mollá-Sirvent},
url = {http://library.iated.org/view/SAVALCALVO2019STU},
doi = {10.21125/inted.2019.0786},
year = {2019},
date = {2019-03-01},
booktitle = {13th International Technology, Education and Development Conference},
pages = {2981--2988},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
|
Fuster-Guilló, Andrés; Azorin-Lopez, Jorge; Jimeno-Morenilla, Antonio; Saval-Calvo, Marcelo; Garcia-Rodriguez, Jose; Mora, Higinio; Pujol, Francisco A; Sanchez-Romero, Jose-Luis; Martínez, Víctor Villena; Garcia-Garcia, Alberto; Sirvent, Rafael Alejandro Mollá Mejora de la motivación en el proceso enseñanza/aprendizaje mediante técnicas de gamificación: caso de estudio en ingeniería informática Artículo de revista En: Investigación e innovación en la Enseñanza Superior. Nuevos contextos, nuevas ideas., pp. 389–397, 2019. @article{Fuster-Guillo2019a,
title = {Mejora de la motivación en el proceso enseñanza/aprendizaje mediante técnicas de gamificación: caso de estudio en ingeniería informática},
author = {Andrés Fuster-Guilló and Jorge Azorin-Lopez and Antonio Jimeno-Morenilla and Marcelo Saval-Calvo and Jose Garcia-Rodriguez and Higinio Mora and Francisco A Pujol and Jose-Luis Sanchez-Romero and Víctor Villena Martínez and Alberto Garcia-Garcia and Rafael Alejandro Mollá Sirvent},
year = {2019},
date = {2019-01-01},
journal = {Investigación e innovación en la Enseñanza Superior. Nuevos contextos, nuevas ideas.},
pages = {389--397},
abstract = {Las técnicas de gamificación trasladan a los entornos educativos la mecánica de los juegos con el objetivo de mejorar el proceso enseñanza aprendizaje. Se busca incentivar la interacción entre el profesor y el estudiante para aumentar la motivación repercutiendo en una mejora de la capacidad de asimilación de conocimientos y adquisición de habilidades. Son conocidas las técnicas basadas en recompensas al usuario (puntos, niveles, premios, desafíos, concursos), así como numerosas las herramientas y plataformas que facilitan la incorporación de la gamificación en el aula. Existen diversos trabajos que miden la incidencia de distintas estrategias de gamificación en la motivación y satisfacción del estudiante, pero es reconocida la carencia de suficientes estudios empíricos que avalúen el impacto en el rendimiento académico de forma contrastada. Este trabajo aporta un estudio de los beneficios que genera el uso de técnicas de gamificación en el contexto de las ingenierías informáticas. Se propone un doble método de gamificación para las sesiones teóricas y prácticas. En el caso de las sesiones teóricas se utilizan cuestionarios interactivos “quizzes”. La estrategia de gamificación de las sesiones prácticas se basa en concursos en formato “hackathon”. Se aporta un análisis del impacto de la actuación en la motivación y grado de satisfacción del estudiante, así como en los resultados de aprendizaje que finalmente se alcanzan.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Las técnicas de gamificación trasladan a los entornos educativos la mecánica de los juegos con el objetivo de mejorar el proceso enseñanza aprendizaje. Se busca incentivar la interacción entre el profesor y el estudiante para aumentar la motivación repercutiendo en una mejora de la capacidad de asimilación de conocimientos y adquisición de habilidades. Son conocidas las técnicas basadas en recompensas al usuario (puntos, niveles, premios, desafíos, concursos), así como numerosas las herramientas y plataformas que facilitan la incorporación de la gamificación en el aula. Existen diversos trabajos que miden la incidencia de distintas estrategias de gamificación en la motivación y satisfacción del estudiante, pero es reconocida la carencia de suficientes estudios empíricos que avalúen el impacto en el rendimiento académico de forma contrastada. Este trabajo aporta un estudio de los beneficios que genera el uso de técnicas de gamificación en el contexto de las ingenierías informáticas. Se propone un doble método de gamificación para las sesiones teóricas y prácticas. En el caso de las sesiones teóricas se utilizan cuestionarios interactivos “quizzes”. La estrategia de gamificación de las sesiones prácticas se basa en concursos en formato “hackathon”. Se aporta un análisis del impacto de la actuación en la motivación y grado de satisfacción del estudiante, así como en los resultados de aprendizaje que finalmente se alcanzan. |
2018
|
Orts-Escolano, Sergio; Garcia-Rodriguez, Jose; Cazorla, Miguel; Morell, Vicente; Azorin, Jorge; Saval, Marcelo; Garcia-Garcia, Alberto; Villena, Victor Bioinspired point cloud representation: 3D object tracking Artículo de revista En: Neural Computing and Applications, vol. 29, no. 9, pp. 663–672, 2018, ISSN: 0941-0643. @article{Orts-Escolano2016,
title = {Bioinspired point cloud representation: 3D object tracking},
author = {Sergio Orts-Escolano and Jose Garcia-Rodriguez and Miguel Cazorla and Vicente Morell and Jorge Azorin and Marcelo Saval and Alberto Garcia-Garcia and Victor Villena},
url = {http://link.springer.com/10.1007/s00521-016-2585-0},
doi = {10.1007/s00521-016-2585-0},
issn = {0941-0643},
year = {2018},
date = {2018-05-01},
journal = {Neural Computing and Applications},
volume = {29},
number = {9},
pages = {663--672},
abstract = {The problem of processing point cloud sequences is considered in this work. In particular, a system that represents and tracks objects in dynamic scenes acquired using low-cost sensors such as the Kinect is presented. An efficient neural network-based approach is proposed to represent and estimate the motion of 3D objects. This system addresses multiple computer vision tasks such as object segmentation, representation, motion analysis and tracking. The use of a neural network allows the unsupervised estimation of motion and the representation of objects in the scene. This proposal avoids the problem of finding corresponding features while tracking moving objects. A set of experiments are presented that demonstrate the validity of our method to track 3D objects. Moreover, an optimization strategy is applied to achieve real-time processing rates. Favorable results are presented demonstrating the capabilities of the GNG-based algorithm for this task.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The problem of processing point cloud sequences is considered in this work. In particular, a system that represents and tracks objects in dynamic scenes acquired using low-cost sensors such as the Kinect is presented. An efficient neural network-based approach is proposed to represent and estimate the motion of 3D objects. This system addresses multiple computer vision tasks such as object segmentation, representation, motion analysis and tracking. The use of a neural network allows the unsupervised estimation of motion and the representation of objects in the scene. This proposal avoids the problem of finding corresponding features while tracking moving objects. A set of experiments are presented that demonstrate the validity of our method to track 3D objects. Moreover, an optimization strategy is applied to achieve real-time processing rates. Favorable results are presented demonstrating the capabilities of the GNG-based algorithm for this task. |
Saval-Calvo, Marcelo; Azorin-Lopez, Jorge; Fuster-Guillo, Andres; Villena-Martinez, Victor; Fisher, Robert B 3D non-rigid registration using color: Color Coherent Point Drift Artículo de revista En: Computer Vision and Image Understanding, vol. 169, pp. 119–135, 2018, ISSN: 10773142. @article{Saval-Calvo2018,
title = {3D non-rigid registration using color: Color Coherent Point Drift},
author = {Marcelo Saval-Calvo and Jorge Azorin-Lopez and Andres Fuster-Guillo and Victor Villena-Martinez and Robert B Fisher},
url = {http://linkinghub.elsevier.com/retrieve/pii/S1077314218300080 https://linkinghub.elsevier.com/retrieve/pii/S1077314218300080},
doi = {10.1016/j.cviu.2018.01.008},
issn = {10773142},
year = {2018},
date = {2018-04-01},
journal = {Computer Vision and Image Understanding},
volume = {169},
pages = {119--135},
abstract = {Research into object deformations using computer vision techniques has been under intense study in recent years. A widely used technique is 3D non-rigid registration to estimate the transformation between two instances of a deforming structure. Despite many previous developments on this topic, it remains a challenging problem. In this paper we propose a novel approach to non-rigid registration combining two data spaces in order to robustly calculate the correspondences and transformation between two data sets. In particular, we use point color as well as 3D location as these are the common outputs of RGB-D cameras. We have propose the Color Coherent Point Drift (CCPD) algorithm (an extension of the CPD method (Myronenko and Song, 2010)). Evaluation is performed using synthetic and real data. The synthetic data includes easy shapes that allow evaluation of the effect of noise, outliers and missing data. Moreover, an evaluation of realistic figures obtained using Blensor is carried out. Real data acquired using a general purpose Primesense Carmine sensor is used to validate the CCPD for real shapes. For all tests, the proposed method is compared to the original CPD showing better results in registration accuracy in most cases.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Research into object deformations using computer vision techniques has been under intense study in recent years. A widely used technique is 3D non-rigid registration to estimate the transformation between two instances of a deforming structure. Despite many previous developments on this topic, it remains a challenging problem. In this paper we propose a novel approach to non-rigid registration combining two data spaces in order to robustly calculate the correspondences and transformation between two data sets. In particular, we use point color as well as 3D location as these are the common outputs of RGB-D cameras. We have propose the Color Coherent Point Drift (CCPD) algorithm (an extension of the CPD method (Myronenko and Song, 2010)). Evaluation is performed using synthetic and real data. The synthetic data includes easy shapes that allow evaluation of the effect of noise, outliers and missing data. Moreover, an evaluation of realistic figures obtained using Blensor is carried out. Real data acquired using a general purpose Primesense Carmine sensor is used to validate the CCPD for real shapes. For all tests, the proposed method is compared to the original CPD showing better results in registration accuracy in most cases. |
Morell-Gimenez, Vicente; Saval-Calvo, Marcelo; Villena-Martinez, Victor; Azorin-Lopez, Jorge; Garcia-Rodriguez, Jose; Cazorla, Miguel; Orts-Escolano, Sergio; Fuster-Guillo, Andres A Survey of 3D Rigid Registration Methods for RGB-D Cameras Parte de obra colectiva En: Advancements in computer vision and Image Processing, pp. 74–98, 2018, ISBN: 9781522556282. @incollection{Morell-Gimenez2018,
title = {A Survey of 3D Rigid Registration Methods for RGB-D Cameras},
author = {Vicente Morell-Gimenez and Marcelo Saval-Calvo and Victor Villena-Martinez and Jorge Azorin-Lopez and Jose Garcia-Rodriguez and Miguel Cazorla and Sergio Orts-Escolano and Andres Fuster-Guillo},
url = {http://services.igi-global.com/resolvedoi/resolve.aspx?doi=10.4018/978-1-5225-5628-2.ch004},
doi = {10.4018/978-1-5225-5628-2.ch004},
isbn = {9781522556282},
year = {2018},
date = {2018-01-01},
booktitle = {Advancements in computer vision and Image Processing},
pages = {74--98},
abstract = {egistration of multiple 3D data sets is a fundamental problem in many areas. Many researches and applications are using low-cost RGB-D sensors for 3D data acquisition. In general terms, the registration problem tries to find a transformation between two coordinate systems that better aligns the point sets. In order to review and describe the state-of-the-art of the rigid registration approaches, the authors decided to classify methods in coarse and fine. Due to the high variety of methods, they have made a study of the registration techniques, which could use RGB-D sensors in static scenarios. This chapter covers most of the expected aspects to consider when a registration technique has to be used with RGB-D sensors. Moreover, in order to establish a taxonomy of the different methods, the authors have classified those using different characteristics. As a result, they present a classification that aims to be a guide to help the researchers or practitioners to select a method based on the requirements of a specific registration problem.},
keywords = {},
pubstate = {published},
tppubtype = {incollection}
}
egistration of multiple 3D data sets is a fundamental problem in many areas. Many researches and applications are using low-cost RGB-D sensors for 3D data acquisition. In general terms, the registration problem tries to find a transformation between two coordinate systems that better aligns the point sets. In order to review and describe the state-of-the-art of the rigid registration approaches, the authors decided to classify methods in coarse and fine. Due to the high variety of methods, they have made a study of the registration techniques, which could use RGB-D sensors in static scenarios. This chapter covers most of the expected aspects to consider when a registration technique has to be used with RGB-D sensors. Moreover, in order to establish a taxonomy of the different methods, the authors have classified those using different characteristics. As a result, they present a classification that aims to be a guide to help the researchers or practitioners to select a method based on the requirements of a specific registration problem. |
2017
|
Azorin-Lopez, Jorge; Saval-Calvo, Marcelo; Fuster-Guillo, Andres; Garcia-Rodriguez, Jose; Mora-Mora, Higinio Constrained self-organizing feature map to preserve feature extraction topology Artículo de revista En: Neural Computing and Applications, vol. 28, no. S1, pp. 439–459, 2017, ISSN: 0941-0643. @article{Azorin-Lopez2016,
title = {Constrained self-organizing feature map to preserve feature extraction topology},
author = {Jorge Azorin-Lopez and Marcelo Saval-Calvo and Andres Fuster-Guillo and Jose Garcia-Rodriguez and Higinio Mora-Mora},
url = {http://link.springer.com/10.1007/s00521-016-2346-0},
doi = {10.1007/s00521-016-2346-0},
issn = {0941-0643},
year = {2017},
date = {2017-12-01},
journal = {Neural Computing and Applications},
volume = {28},
number = {S1},
pages = {439--459},
abstract = {In many classification problems, it is necessary to consider the specific location of an n-dimensional space from which features have been calculated. For example, considering the location of features extracted from specific areas of a two-dimensional space, as an image, could improve the understanding of a scene for a video surveillance system. In the same way, the same features extracted from different locations could mean different actions for a 3D HCI system. In this paper, we present a self-organizing feature map able to preserve the topology of locations of an n-dimensional space in which the vector of features have been extracted. The main contribution is to implicitly preserving the topology of the original space because considering the locations of the extracted features and their topology could ease the solution to certain problems. Specifically, the paper proposes the n-dimensional constrained self-organizing map preserving the input topology (nD-SOM-PINT). Features in adjacent areas of the n-dimensional space, used to extract the feature vectors, are explicitly in adjacent areas of the nD-SOM-PINT constraining the neural network structure and learning. As a study case, the neural network has been instantiate to represent and classify features as trajectories extracted from a sequence of images into a high level of semantic understanding. Experiments have been thoroughly carried out using the CAVIAR datasets (Corridor, Frontal and Inria) taken into account the global behaviour of an individual in order to validate the ability to preserve the topology of the two-dimensional space to obtain high-performance classification for trajectory classification in contrast of non-considering the location of features. Moreover, a brief example has been included to focus on validate the nD-SOM-PINT proposal in other domain than the individual trajectory. Results confirm the high accuracy of the nD-SOM-PINT outperforming previous methods aimed to classify the same datasets.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
In many classification problems, it is necessary to consider the specific location of an n-dimensional space from which features have been calculated. For example, considering the location of features extracted from specific areas of a two-dimensional space, as an image, could improve the understanding of a scene for a video surveillance system. In the same way, the same features extracted from different locations could mean different actions for a 3D HCI system. In this paper, we present a self-organizing feature map able to preserve the topology of locations of an n-dimensional space in which the vector of features have been extracted. The main contribution is to implicitly preserving the topology of the original space because considering the locations of the extracted features and their topology could ease the solution to certain problems. Specifically, the paper proposes the n-dimensional constrained self-organizing map preserving the input topology (nD-SOM-PINT). Features in adjacent areas of the n-dimensional space, used to extract the feature vectors, are explicitly in adjacent areas of the nD-SOM-PINT constraining the neural network structure and learning. As a study case, the neural network has been instantiate to represent and classify features as trajectories extracted from a sequence of images into a high level of semantic understanding. Experiments have been thoroughly carried out using the CAVIAR datasets (Corridor, Frontal and Inria) taken into account the global behaviour of an individual in order to validate the ability to preserve the topology of the two-dimensional space to obtain high-performance classification for trajectory classification in contrast of non-considering the location of features. Moreover, a brief example has been included to focus on validate the nD-SOM-PINT proposal in other domain than the individual trajectory. Results confirm the high accuracy of the nD-SOM-PINT outperforming previous methods aimed to classify the same datasets. |
Borja, Luis Felipe; Azorin-Lopez, Jorge; Saval-Calvo, Marcelo A Compilation of Methods and Datasets for Group and Crowd Action Recognition Artículo de revista En: International Journal of Computer Vision and Image Processing, vol. 7, no. 3, pp. 40–53, 2017, ISSN: 2155-6997. @article{Borja2017a,
title = {A Compilation of Methods and Datasets for Group and Crowd Action Recognition},
author = {Luis Felipe Borja and Jorge Azorin-Lopez and Marcelo Saval-Calvo},
url = {http://services.igi-global.com/resolvedoi/resolve.aspx?doi=10.4018/IJCVIP.2017070104},
doi = {10.4018/IJCVIP.2017070104},
issn = {2155-6997},
year = {2017},
date = {2017-07-01},
journal = {International Journal of Computer Vision and Image Processing},
volume = {7},
number = {3},
pages = {40--53},
abstract = {The human behaviour analysis has been a subject of study in various fields of science (e.g. sociology, psychology, computer science). Specifically, the automated understanding of the behaviour of both individuals and groups remains a very challenging problem from the sensor systems to artificial intelligence techniques. Being aware of the extent of the topic, the objective of this paper is to review the state of the art focusing on machine learning techniques and computer vision as sensor system to the artificial intelligence techniques. Moreover, a lack of review comparing the level of abstraction in terms of activities duration is found in the literature. In this paper, a review of the methods and techniques based on machine learning to classify group behaviour in sequence of images is presented. The review takes into account the different levels of understanding and the number of people in the group.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The human behaviour analysis has been a subject of study in various fields of science (e.g. sociology, psychology, computer science). Specifically, the automated understanding of the behaviour of both individuals and groups remains a very challenging problem from the sensor systems to artificial intelligence techniques. Being aware of the extent of the topic, the objective of this paper is to review the state of the art focusing on machine learning techniques and computer vision as sensor system to the artificial intelligence techniques. Moreover, a lack of review comparing the level of abstraction in terms of activities duration is found in the literature. In this paper, a review of the methods and techniques based on machine learning to classify group behaviour in sequence of images is presented. The review takes into account the different levels of understanding and the number of people in the group. |
Borja, Luis Felipe; Azorin-Lopez, Jorge; Saval-Calvo, Marcelo A Compilation of Methods and Datasets for Group and Crowd Action Recognition Artículo de revista En: International Journal of Computer Vision and Image Processing, vol. 7, no. 3, pp. 40–53, 2017, ISSN: 2155-6997. @article{Borja2017,
title = {A Compilation of Methods and Datasets for Group and Crowd Action Recognition},
author = {Luis Felipe Borja and Jorge Azorin-Lopez and Marcelo Saval-Calvo},
url = {http://services.igi-global.com/resolvedoi/resolve.aspx?doi=10.4018/IJCVIP.2017070104},
doi = {10.4018/IJCVIP.2017070104},
issn = {2155-6997},
year = {2017},
date = {2017-07-01},
journal = {International Journal of Computer Vision and Image Processing},
volume = {7},
number = {3},
pages = {40--53},
abstract = {The human behaviour analysis has been a subject of study in various fields of science (e.g. sociology, psychology, computer science). Specifically, the automated understanding of the behaviour of both individuals and groups remains a very challenging problem from the sensor systems to artificial intelligence techniques. Being aware of the extent of the topic, the objective of this paper is to review the state of the art focusing on machine learning techniques and computer vision as sensor system to the artificial intelligence techniques. Moreover, a lack of review comparing the level of abstraction in terms of activities duration is found in the literature. In this paper, a review of the methods and techniques based on machine learning to classify group behaviour in sequence of images is presented. The review takes into account the different levels of understanding and the number of people in the group.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The human behaviour analysis has been a subject of study in various fields of science (e.g. sociology, psychology, computer science). Specifically, the automated understanding of the behaviour of both individuals and groups remains a very challenging problem from the sensor systems to artificial intelligence techniques. Being aware of the extent of the topic, the objective of this paper is to review the state of the art focusing on machine learning techniques and computer vision as sensor system to the artificial intelligence techniques. Moreover, a lack of review comparing the level of abstraction in terms of activities duration is found in the literature. In this paper, a review of the methods and techniques based on machine learning to classify group behaviour in sequence of images is presented. The review takes into account the different levels of understanding and the number of people in the group. |
Villena-Martinez, Victor; Fuster-Guillo, Andres; Saval-Calvo, Marcelo; Azorin-Lopez, Jorge An Iterative Method for 3D Body Registration Using a Single RGB-D Sensor Artículo de revista En: International Journal of Computer Vision and Image Processing, vol. 7, no. 3, pp. 26–39, 2017, ISSN: 2155-6997. @article{Villena-Martinez2017,
title = {An Iterative Method for 3D Body Registration Using a Single RGB-D Sensor},
author = {Victor Villena-Martinez and Andres Fuster-Guillo and Marcelo Saval-Calvo and Jorge Azorin-Lopez},
url = {http://services.igi-global.com/resolvedoi/resolve.aspx?doi=10.4018/IJCVIP.2017070103},
doi = {10.4018/IJCVIP.2017070103},
issn = {2155-6997},
year = {2017},
date = {2017-07-01},
journal = {International Journal of Computer Vision and Image Processing},
volume = {7},
number = {3},
pages = {26--39},
abstract = {In this paper, the problem of 3D body registration using a single RGB-D sensor is approached. It has been guided by three main requirements: low-cost, unconstrained movement and accuracy. In order to fit them, an iterative registration method for accurately aligning data from single RGB-D sensor is proposed. The data is acquired while a person rotates in front of the camera, without the need of any external marker or constraint about its pose. The articulated alignment is carried out in a model-free approach in order to be more consistent with the real data. The iterative method is divided in stages, contributing to each other by the refinement of a specific part of the acquired data. The exploratory results validate the proposed method that is able to feed on itself in each iteration improving the final result by a progressive iteration, with the required precision under the conditions of affordability and unconstrained movement acquisition.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
In this paper, the problem of 3D body registration using a single RGB-D sensor is approached. It has been guided by three main requirements: low-cost, unconstrained movement and accuracy. In order to fit them, an iterative registration method for accurately aligning data from single RGB-D sensor is proposed. The data is acquired while a person rotates in front of the camera, without the need of any external marker or constraint about its pose. The articulated alignment is carried out in a model-free approach in order to be more consistent with the real data. The iterative method is divided in stages, contributing to each other by the refinement of a specific part of the acquired data. The exploratory results validate the proposed method that is able to feed on itself in each iteration improving the final result by a progressive iteration, with the required precision under the conditions of affordability and unconstrained movement acquisition. |
Azorin-Lopez, Jorge; Fuster-Guillo, Andres; Saval-Calvo, Marcelo; Mora-MoraJ, Higinio; Garcia-Chamizo, Juan Manuel A Novel Active Imaging Model to Design Visual Systems: A Case of Inspection System for Specular Surfaces Artículo de revista En: Sensors, vol. 17, no. 7, pp. 1466, 2017, ISSN: 1424-8220. @article{Azorin-Lopez2017a,
title = {A Novel Active Imaging Model to Design Visual Systems: A Case of Inspection System for Specular Surfaces},
author = {Jorge Azorin-Lopez and Andres Fuster-Guillo and Marcelo Saval-Calvo and Higinio Mora-MoraJ and Juan Manuel Garcia-Chamizo},
url = {http://www.mdpi.com/1424-8220/17/7/1466},
doi = {10.3390/s17071466},
issn = {1424-8220},
year = {2017},
date = {2017-06-01},
journal = {Sensors},
volume = {17},
number = {7},
pages = {1466},
abstract = {The use of visual information is a very well known input from different kinds of sensors. However, most of the perception problems are individually modeled and tackled. It is necessary to provide a general imaging model that allows us to parametrize different input systems as well as their problems and possible solutions. In this paper, we present an active vision model considering the imaging system as a whole (including camera, lighting system, object to be perceived) in order to propose solutions to automated visual systems that present problems that we perceive. As a concrete case study, we instantiate the model in a real application and still challenging problem: automated visual inspection. It is one of the most used quality control systems to detect defects on manufactured objects. However, it presents problems for specular products. We model these perception problems taking into account environmental conditions and camera parameters that allow a system to properly perceive the specific object characteristics to determine defects on surfaces. The validation of the model has been carried out using simulations providing an efficient way to perform a large set of tests (different environment conditions and camera parameters) as a previous step of experimentation in real manufacturing environments, which more complex in terms of instrumentation and more expensive. Results prove the success of the model application adjusting scale, viewpoint and lighting conditions to detect structural and color defects on specular surfaces.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The use of visual information is a very well known input from different kinds of sensors. However, most of the perception problems are individually modeled and tackled. It is necessary to provide a general imaging model that allows us to parametrize different input systems as well as their problems and possible solutions. In this paper, we present an active vision model considering the imaging system as a whole (including camera, lighting system, object to be perceived) in order to propose solutions to automated visual systems that present problems that we perceive. As a concrete case study, we instantiate the model in a real application and still challenging problem: automated visual inspection. It is one of the most used quality control systems to detect defects on manufactured objects. However, it presents problems for specular products. We model these perception problems taking into account environmental conditions and camera parameters that allow a system to properly perceive the specific object characteristics to determine defects on surfaces. The validation of the model has been carried out using simulations providing an efficient way to perform a large set of tests (different environment conditions and camera parameters) as a previous step of experimentation in real manufacturing environments, which more complex in terms of instrumentation and more expensive. Results prove the success of the model application adjusting scale, viewpoint and lighting conditions to detect structural and color defects on specular surfaces. |
Saval-Calvo, Marcelo; Azorin-Lopez, Jorge; Fuster-Guillo, Andres; Garcia-Rodriguez, Jose; Orts-Escolano, Sergio; Garcia-Garcia, Alberto Evaluation of sampling method effects in 3D non-rigid registration Artículo de revista En: Neural Computing and Applications, vol. 28, no. 5, pp. 953–967, 2017, ISSN: 0941-0643. @article{Saval-Calvo2016,
title = {Evaluation of sampling method effects in 3D non-rigid registration},
author = {Marcelo Saval-Calvo and Jorge Azorin-Lopez and Andres Fuster-Guillo and Jose Garcia-Rodriguez and Sergio Orts-Escolano and Alberto Garcia-Garcia},
url = {http://link.springer.com/10.1007/s00521-016-2258-z},
doi = {10.1007/s00521-016-2258-z},
issn = {0941-0643},
year = {2017},
date = {2017-05-01},
journal = {Neural Computing and Applications},
volume = {28},
number = {5},
pages = {953--967},
abstract = {Since the beginning of 3D computer vision problems, the use of techniques to reduce the data to make it treatable preserving the important aspects of the scene has been necessary. Currently, with the new low-cost RGB-D sensors, which provide a stream of color and 3D data of approximately 30 frames per second, this is getting more relevance. Many applications make use of these sensors and need a preprocessing to downsample the data in order to either reduce the processing time or improve the data (e.g., reducing noise or enhancing the important features). In this paper, we present a comparison of different downsampling techniques which are based on different principles. Concretely, five different downsampling methods are included: a bilinear-based method, a normal-based, a color-based, a combination of the normal and color-based samplings, and a growing neural gas (GNG)-based approach. For the comparison, two different models have been used acquired with the Blensor software. Moreover, to evaluate the effect of the downsampling in a real application, a 3D non-rigid registration is performed with the data sampled. From the experimentation we can conclude that depending on the purpose of the application some kernels of the sampling methods can improve drastically the results. Bilinear- and GNG-based methods provide homogeneous point clouds, but color-based and normal-based provide datasets with higher density of points in areas with specific features. In the non-rigid application, if a color-based sampled point cloud is used, it is possible to properly register two datasets for cases where intensity data are relevant in the model and outperform the results if only a homogeneous sampling is used.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Since the beginning of 3D computer vision problems, the use of techniques to reduce the data to make it treatable preserving the important aspects of the scene has been necessary. Currently, with the new low-cost RGB-D sensors, which provide a stream of color and 3D data of approximately 30 frames per second, this is getting more relevance. Many applications make use of these sensors and need a preprocessing to downsample the data in order to either reduce the processing time or improve the data (e.g., reducing noise or enhancing the important features). In this paper, we present a comparison of different downsampling techniques which are based on different principles. Concretely, five different downsampling methods are included: a bilinear-based method, a normal-based, a color-based, a combination of the normal and color-based samplings, and a growing neural gas (GNG)-based approach. For the comparison, two different models have been used acquired with the Blensor software. Moreover, to evaluate the effect of the downsampling in a real application, a 3D non-rigid registration is performed with the data sampled. From the experimentation we can conclude that depending on the purpose of the application some kernels of the sampling methods can improve drastically the results. Bilinear- and GNG-based methods provide homogeneous point clouds, but color-based and normal-based provide datasets with higher density of points in areas with specific features. In the non-rigid application, if a color-based sampled point cloud is used, it is possible to properly register two datasets for cases where intensity data are relevant in the model and outperform the results if only a homogeneous sampling is used. |
Garcia-Garcia, Alberto; Orts-Escolano, Sergio; Oprea, Sergiu; Garcia-Rodriguez, Jose; Azorin-Lopez, Jorge; Saval-Calvo, Marcelo; Cazorla, Miguel Multi-sensor 3D object dataset for object recognition with full pose estimation Artículo de revista En: Neural Computing and Applications, vol. 28, no. 5, pp. 941–952, 2017, ISSN: 0941-0643. @article{Garcia-Garcia2016,
title = {Multi-sensor 3D object dataset for object recognition with full pose estimation},
author = {Alberto Garcia-Garcia and Sergio Orts-Escolano and Sergiu Oprea and Jose Garcia-Rodriguez and Jorge Azorin-Lopez and Marcelo Saval-Calvo and Miguel Cazorla},
url = {http://link.springer.com/10.1007/s00521-016-2224-9},
doi = {10.1007/s00521-016-2224-9},
issn = {0941-0643},
year = {2017},
date = {2017-05-01},
journal = {Neural Computing and Applications},
volume = {28},
number = {5},
pages = {941--952},
abstract = {In this work, we propose a new dataset for 3D object recognition using the new high-resolution Kinect V2 sensor and some other popular low-cost devices like PrimeSense Carmine. Since most already existing datasets for 3D object recognition lack some features such as 3D pose information about objects in the scene, per pixel segmentation or level of occlusion, we propose a new one combining all this information in a single dataset that can be used to validate existing and new 3D object recognition algorithms. Moreover, with the advent of the new Kinect V2 sensor we are able to provide high-resolution data for RGB and depth information using a single sensor, whereas other datasets had to combine multiple sensors. In addition, we will also provide semiautomatic segmentation and semantic labels about the different parts of the objects so that the dataset could be used for testing robot grasping and scene labeling systems as well as for object recognition.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
In this work, we propose a new dataset for 3D object recognition using the new high-resolution Kinect V2 sensor and some other popular low-cost devices like PrimeSense Carmine. Since most already existing datasets for 3D object recognition lack some features such as 3D pose information about objects in the scene, per pixel segmentation or level of occlusion, we propose a new one combining all this information in a single dataset that can be used to validate existing and new 3D object recognition algorithms. Moreover, with the advent of the new Kinect V2 sensor we are able to provide high-resolution data for RGB and depth information using a single sensor, whereas other datasets had to combine multiple sensors. In addition, we will also provide semiautomatic segmentation and semantic labels about the different parts of the objects so that the dataset could be used for testing robot grasping and scene labeling systems as well as for object recognition. |
Azorin-Lopez, Jorge; Garcia-Rodriguez, Jose; Jimeno-Morenilla, Antonio; Mora-Mora, Higinio; Pujol-Lopez, Francisco; Sanchez-Romero, Jose Luis; Fuster-Guillo, Andres; Saval-Calvo, Marcelo; Garcia-Garcia, Alberto TEACHING AND LEARNING METHODOLOGIES FOR PRACTICAL ASSIGNMENTS APPLIED TO SMALL GROUPS OF SPANISH TOP-MARK STUDENTS Artículo en actas En: 11th International Technology, Education and Development Conference, pp. 6825–6834, 2017. @inproceedings{Azorin-Lopez2017b,
title = {TEACHING AND LEARNING METHODOLOGIES FOR PRACTICAL ASSIGNMENTS APPLIED TO SMALL GROUPS OF SPANISH TOP-MARK STUDENTS},
author = {Jorge Azorin-Lopez and Jose Garcia-Rodriguez and Antonio Jimeno-Morenilla and Higinio Mora-Mora and Francisco Pujol-Lopez and Jose Luis Sanchez-Romero and Andres Fuster-Guillo and Marcelo Saval-Calvo and Alberto Garcia-Garcia},
url = {http://library.iated.org/view/AZORINLOPEZ2017TEA},
doi = {10.21125/inted.2017.1586},
year = {2017},
date = {2017-03-01},
booktitle = {11th International Technology, Education and Development Conference},
pages = {6825--6834},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
|
Azorin-Lopez, Jorge; Garcia-Rodriguez, Jose; Jimeno-Morenilla, Antonio; Mora-Mora, Higinio; Pujol-Lopez, Francisco; Sanchez-Romero, Jose Luis; Fuster-Guillo, Andres; Saval-Calvo, Marcelo; Garcia-Garcia, Alberto TEACHING AND LEARNING METHODOLOGIES FOR PRACTICAL ASSIGNMENTS APPLIED TO SMALL GROUPS OF SPANISH TOP-MARK STUDENTS Artículo en actas En: 11th International Technology, Education and Development Conference (INTED), pp. 6825–6834, Valencia, 2017. @inproceedings{Azorin-Lopez2017,
title = {TEACHING AND LEARNING METHODOLOGIES FOR PRACTICAL ASSIGNMENTS APPLIED TO SMALL GROUPS OF SPANISH TOP-MARK STUDENTS},
author = {Jorge Azorin-Lopez and Jose Garcia-Rodriguez and Antonio Jimeno-Morenilla and Higinio Mora-Mora and Francisco Pujol-Lopez and Jose Luis Sanchez-Romero and Andres Fuster-Guillo and Marcelo Saval-Calvo and Alberto Garcia-Garcia},
url = {http://library.iated.org/view/AZORINLOPEZ2017TEA},
doi = {10.21125/inted.2017.1586},
year = {2017},
date = {2017-03-01},
booktitle = {11th International Technology, Education and Development Conference (INTED)},
pages = {6825--6834},
address = {Valencia},
abstract = {The incorporation of European universities into the EHEA has made them share a framework in which to develop jointly. This action is part of a higher process called globalization in which the trends marked by technological, economic and industrial development forces universities to play a new role. Individuals and societies will develop successfully only if there is a system of joint professional development. In this context, the skills of teamwork and the use of the English language are fundamental for graduates of EU universities. Teaching in English is one of the challenges that the Spanish university is currently dealing with. The University of Alicante offers, through the groups of High Academic Performance (HAP), part of the teaching of the degrees in English. The main objective of this paper is to show the research carried out on learning methodologies for HAP groups in the field of computer architecture as a case of study. In particular, a project based learning methodology is designed in order to strengthen active participation and teamwork of the students. Moreover, the use of English as the language for teaching and learning the course is, by itself, a breakthrough and a curricular difference competitive advantage that will enable students to compete with certain advantages when they finish the degree. In addition, the promotion of the capacity of effort in the search of solutions that is strengthened in the students has allowed to increase the responsibility in their task of learning. The academic results of this method have been very encouraging, achieving a large number of maximum qualifications.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The incorporation of European universities into the EHEA has made them share a framework in which to develop jointly. This action is part of a higher process called globalization in which the trends marked by technological, economic and industrial development forces universities to play a new role. Individuals and societies will develop successfully only if there is a system of joint professional development. In this context, the skills of teamwork and the use of the English language are fundamental for graduates of EU universities. Teaching in English is one of the challenges that the Spanish university is currently dealing with. The University of Alicante offers, through the groups of High Academic Performance (HAP), part of the teaching of the degrees in English. The main objective of this paper is to show the research carried out on learning methodologies for HAP groups in the field of computer architecture as a case of study. In particular, a project based learning methodology is designed in order to strengthen active participation and teamwork of the students. Moreover, the use of English as the language for teaching and learning the course is, by itself, a breakthrough and a curricular difference competitive advantage that will enable students to compete with certain advantages when they finish the degree. In addition, the promotion of the capacity of effort in the search of solutions that is strengthened in the students has allowed to increase the responsibility in their task of learning. The academic results of this method have been very encouraging, achieving a large number of maximum qualifications. |
Castillo-Secilla, J M; Saval-Calvo, M; Medina-Valdès, L; Cuenca-Asensi, S; Martínez-Álvarez, A; Sánchez, C; Cristóbal, G Autofocus method for automated microscopy using embedded GPUs Artículo de revista En: Biomedical Optics Express, vol. 8, no. 3, pp. 1731, 2017, ISSN: 2156-7085. @article{Castillo-Secilla2017,
title = {Autofocus method for automated microscopy using embedded GPUs},
author = {J M Castillo-Secilla and M Saval-Calvo and L Medina-Valdès and S Cuenca-Asensi and A Martínez-Álvarez and C Sánchez and G Cristóbal},
url = {https://www.osapublishing.org/abstract.cfm?URI=boe-8-3-1731},
doi = {10.1364/BOE.8.001731},
issn = {2156-7085},
year = {2017},
date = {2017-03-01},
journal = {Biomedical Optics Express},
volume = {8},
number = {3},
pages = {1731},
abstract = {In this paper we present a method for autofocusing images of sputum smears taken from a microscope which combines the finding of the optimal focus distance with an algorithm for extending the depth of field (EDoF). Our multifocus fusion method produces an unique image where all the relevant objects of the analyzed scene are well focused, independently to their distance to the sensor. This process is computationally expensive which makes unfeasible its automation using traditional embedded processors. For this purpose a low-cost optimized implementation is proposed using limited resources embedded GPU integrated on cutting-edge NVIDIA system on chip. The extensive tests performed on different sputum smear image sets show the real-time capabilities of our implementation maintaining the quality of the output image.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
In this paper we present a method for autofocusing images of sputum smears taken from a microscope which combines the finding of the optimal focus distance with an algorithm for extending the depth of field (EDoF). Our multifocus fusion method produces an unique image where all the relevant objects of the analyzed scene are well focused, independently to their distance to the sensor. This process is computationally expensive which makes unfeasible its automation using traditional embedded processors. For this purpose a low-cost optimized implementation is proposed using limited resources embedded GPU integrated on cutting-edge NVIDIA system on chip. The extensive tests performed on different sputum smear image sets show the real-time capabilities of our implementation maintaining the quality of the output image. |
Saval-Calvo, Marcelo; Medina-Valdés, Luis; Castillo-Secilla, José; Cuenca-Asensi, Sergio; Alvarez, Antonio Martínezl; Villagrá, Jorge A Review of the Bayesian Occupancy Filter Artículo de revista En: Sensors, vol. 17, no. 2, pp. 344, 2017, ISSN: 1424-8220. @article{Saval-Calvo2017,
title = {A Review of the Bayesian Occupancy Filter},
author = {Marcelo Saval-Calvo and Luis Medina-Valdés and José Castillo-Secilla and Sergio Cuenca-Asensi and Antonio Martínezl Alvarez and Jorge Villagrá},
url = {http://www.mdpi.com/1424-8220/17/2/344},
doi = {10.3390/s17020344},
issn = {1424-8220},
year = {2017},
date = {2017-02-01},
journal = {Sensors},
volume = {17},
number = {2},
pages = {344},
abstract = {Autonomous vehicle systems are currently the object of intense research within scientific and industrial communities; however, many problems remain to be solved. One of the most critical aspects addressed in both autonomous driving and robotics is environment perception, since it consists of the ability to understand the surroundings of the vehicle to estimate risks and make decisions on future movements. In recent years, the Bayesian Occupancy Filter (BOF) method has been developed to evaluate occupancy by tessellation of the environment. A review of the BOF and its variants is presented in this paper. Moreover, we propose a detailed taxonomy where the BOF is decomposed into five progressive layers, from the level closest to the sensor to the highest abstractlevelofriskassessment. Inaddition,wepresentastudyofimplementedusecasestoprovide a practical understanding on the main uses of the BOF and its taxonomy.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Autonomous vehicle systems are currently the object of intense research within scientific and industrial communities; however, many problems remain to be solved. One of the most critical aspects addressed in both autonomous driving and robotics is environment perception, since it consists of the ability to understand the surroundings of the vehicle to estimate risks and make decisions on future movements. In recent years, the Bayesian Occupancy Filter (BOF) method has been developed to evaluate occupancy by tessellation of the environment. A review of the BOF and its variants is presented in this paper. Moreover, we propose a detailed taxonomy where the BOF is decomposed into five progressive layers, from the level closest to the sensor to the highest abstractlevelofriskassessment. Inaddition,wepresentastudyofimplementedusecasestoprovide a practical understanding on the main uses of the BOF and its taxonomy. |
Maniatis, Christos; Saval-calvo, Marcelo; Tylecek, Radim; Fisher, Robert B Best Viewpoint Tracking for Camera Mounted on Robotic Arm with Dynamic Obstacles Artículo en actas En: International Conference on 3D Vision, 2017. @inproceedings{Maniatis2017,
title = {Best Viewpoint Tracking for Camera Mounted on Robotic Arm with Dynamic Obstacles},
author = {Christos Maniatis and Marcelo Saval-calvo and Radim Tylecek and Robert B Fisher},
year = {2017},
date = {2017-01-01},
booktitle = {International Conference on 3D Vision},
abstract = {The problem of finding a next best viewpoint for 3D modeling or scene mapping has been explored in computer vision over the last decade. This paper tackles a similar problem, but with different characteristics. It proposes a method for dynamic next best viewpoint recovery of a target point while avoiding possible occlusions. Since the environment can change, the method has to iteratively find the next best view with a global understanding of the free and occupied parts. We model the problem as a set of possible viewpoints which correspond to the centers of the facets of a virtual tessellated hemisphere covering the scene. Taking into account occlusions, distances between current and future viewpoints, quality of the viewpoint and joint constraints (robot arm joint distances or limits), we evaluate the next best viewpoint. The proposal has been evaluated on 8 different scenarios with different occlusions and a short 3D video sequence to validate its dynamic performance.Collapse},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The problem of finding a next best viewpoint for 3D modeling or scene mapping has been explored in computer vision over the last decade. This paper tackles a similar problem, but with different characteristics. It proposes a method for dynamic next best viewpoint recovery of a target point while avoiding possible occlusions. Since the environment can change, the method has to iteratively find the next best view with a global understanding of the free and occupied parts. We model the problem as a set of possible viewpoints which correspond to the centers of the facets of a virtual tessellated hemisphere covering the scene. Taking into account occlusions, distances between current and future viewpoints, quality of the viewpoint and joint constraints (robot arm joint distances or limits), we evaluate the next best viewpoint. The proposal has been evaluated on 8 different scenarios with different occlusions and a short 3D video sequence to validate its dynamic performance.Collapse |
Villena-Martínez, Víctor; Fuster-Guilló, Andrés; Azorín-López, Jorge; Saval-Calvo, Marcelo; Mora-Pascual, Jeronimo; Garcia-Rodriguez, Jose; Garcia-Garcia, Alberto A Quantitative Comparison of Calibration Methods for RGB-D Sensors Using Different Technologies Artículo de revista En: Sensors, vol. 17, no. 2, pp. 243, 2017, ISSN: 1424-8220. @article{Villena-Martinez2017a,
title = {A Quantitative Comparison of Calibration Methods for RGB-D Sensors Using Different Technologies},
author = {Víctor Villena-Martínez and Andrés Fuster-Guilló and Jorge Azorín-López and Marcelo Saval-Calvo and Jeronimo Mora-Pascual and Jose Garcia-Rodriguez and Alberto Garcia-Garcia},
url = {http://www.mdpi.com/1424-8220/17/2/243},
doi = {10.3390/s17020243},
issn = {1424-8220},
year = {2017},
date = {2017-01-01},
journal = {Sensors},
volume = {17},
number = {2},
pages = {243},
abstract = {RGB-D (Red Green Blue and Depth) sensors are devices that can provide color and depth information from a scene at the same time. Recently, they have been widely used in many solutions due to their commercial growth from the entertainment market to many diverse areas (e.g., robotics, CAD, etc.). In the research community, these devices have had good uptake due to their acceptable levelofaccuracyformanyapplicationsandtheirlowcost,butinsomecases,theyworkatthelimitof their sensitivity, near to the minimum feature size that can be perceived. For this reason, calibration processes are critical in order to increase their accuracy and enable them to meet the requirements of such kinds of applications. To the best of our knowledge, there is not a comparative study of calibration algorithms evaluating its results in multiple RGB-D sensors. Specifically, in this paper, a comparison of the three most used calibration methods have been applied to three different RGB-D sensors based on structured light and time-of-flight. The comparison of methods has been carried out by a set of experiments to evaluate the accuracy of depth measurements. Additionally, an object reconstruction application has been used as example of an application for which the sensor works at the limit of its sensitivity. The obtained results of reconstruction have been evaluated through visual inspection and quantitative measurements.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
RGB-D (Red Green Blue and Depth) sensors are devices that can provide color and depth information from a scene at the same time. Recently, they have been widely used in many solutions due to their commercial growth from the entertainment market to many diverse areas (e.g., robotics, CAD, etc.). In the research community, these devices have had good uptake due to their acceptable levelofaccuracyformanyapplicationsandtheirlowcost,butinsomecases,theyworkatthelimitof their sensitivity, near to the minimum feature size that can be perceived. For this reason, calibration processes are critical in order to increase their accuracy and enable them to meet the requirements of such kinds of applications. To the best of our knowledge, there is not a comparative study of calibration algorithms evaluating its results in multiple RGB-D sensors. Specifically, in this paper, a comparison of the three most used calibration methods have been applied to three different RGB-D sensors based on structured light and time-of-flight. The comparison of methods has been carried out by a set of experiments to evaluate the accuracy of depth measurements. Additionally, an object reconstruction application has been used as example of an application for which the sensor works at the limit of its sensitivity. The obtained results of reconstruction have been evaluated through visual inspection and quantitative measurements. |
2016
|
Azorin-Lopez, Jorge; Saval-Calvo, Marcelo; Fuster-Guillo, Andres; Garcia-Rodriguez, Jose; Cazorla, Miguel; Signes-Pont, Maria Teresa Group activity description and recognition based on trajectory analysis and neural networks Artículo en actas En: 2016 International Joint Conference on Neural Networks (IJCNN), pp. 1585–1592, IEEE, 2016, ISBN: 978-1-5090-0620-5. @inproceedings{Azorin-Lopez2016b,
title = {Group activity description and recognition based on trajectory analysis and neural networks},
author = {Jorge Azorin-Lopez and Marcelo Saval-Calvo and Andres Fuster-Guillo and Jose Garcia-Rodriguez and Miguel Cazorla and Maria Teresa Signes-Pont},
url = {http://ieeexplore.ieee.org/document/7727387/},
doi = {10.1109/IJCNN.2016.7727387},
isbn = {978-1-5090-0620-5},
year = {2016},
date = {2016-07-01},
booktitle = {2016 International Joint Conference on Neural Networks (IJCNN)},
pages = {1585--1592},
publisher = {IEEE},
abstract = {The recognition of group activities using computer vision and pattern recognition methods has been, and still remains, a challenging problem. Most of the research on human behaviour has been focused on recognizing individual issues from actions to behaviours. However, the analysis and recognition of group activities, the relationships of different groups in the scene and the interaction of the individuals in the group is still considered an open problem. This paper proposes a novel representation method to analyse and recognise group activities, called Group Activity Descriptor Vector (GADV). It is calculated from the trajectory described by the group and by the individuals who form it. Specifically, the GADV describes three different components: the trajectory followed by the group, the coherence of the individual trajectories in the group and, finally, the movement relationships among different groups in the scene. The trajectory analysis allows a simple high level understanding of complex groups activities. The GADV representation has been evaluated with different self-organizing neural networks using Behave and Caviar dataset sequences obtaining great accuracy in the recognition of the group activities, outperforming the state of the art methods.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The recognition of group activities using computer vision and pattern recognition methods has been, and still remains, a challenging problem. Most of the research on human behaviour has been focused on recognizing individual issues from actions to behaviours. However, the analysis and recognition of group activities, the relationships of different groups in the scene and the interaction of the individuals in the group is still considered an open problem. This paper proposes a novel representation method to analyse and recognise group activities, called Group Activity Descriptor Vector (GADV). It is calculated from the trajectory described by the group and by the individuals who form it. Specifically, the GADV describes three different components: the trajectory followed by the group, the coherence of the individual trajectories in the group and, finally, the movement relationships among different groups in the scene. The trajectory analysis allows a simple high level understanding of complex groups activities. The GADV representation has been evaluated with different self-organizing neural networks using Behave and Caviar dataset sequences obtaining great accuracy in the recognition of the group activities, outperforming the state of the art methods. |
Azorin-Lopez, Jorge; Saval-Calvo, Marcelo; Fuster-Guillo, Andres; Garcia-Rodriguez, Jose A Novel Prediction Method for Early Recognition of Global Human Behaviour in Image Sequences Artículo de revista En: Neural Processing Letters, vol. 43, no. 2, pp. 363–387, 2016, ISSN: 1370-4621. @article{Azorin-Lopez2016c,
title = {A Novel Prediction Method for Early Recognition of Global Human Behaviour in Image Sequences},
author = {Jorge Azorin-Lopez and Marcelo Saval-Calvo and Andres Fuster-Guillo and Jose Garcia-Rodriguez},
url = {http://link.springer.com/10.1007/s11063-015-9412-y},
doi = {10.1007/s11063-015-9412-y},
issn = {1370-4621},
year = {2016},
date = {2016-04-01},
journal = {Neural Processing Letters},
volume = {43},
number = {2},
pages = {363--387},
abstract = {Human behaviour recognition has been, and still remains, a challenging problem that involves different areas of computational intelligence. The automated understanding of people activities from video sequences is an open research topic in which the computer vision and pattern recognition areas have made big efforts. In this paper, the problem is studied from a prediction point of view. We propose a novel method able to early detect behaviour using a small portion of the input, in addition to the capabilities of it to predict behaviour from new inputs. Specifically, we propose a predictive method based on a simple representation of trajectories of a person in the scene which allows a high level understanding of the global human behaviour. The representation of the trajectory is used as a descriptor of the activity of the individual. The descriptors are used as a cue of a classification stage for pattern recognition purposes. Classifiers are trained using the trajectory representation of the complete sequence. However, partial sequences are processed to evaluate the early prediction capabilities having a specific observation time of the scene. The experiments have been carried out using the three different dataset of the CAVIAR database taken into account the behaviour of an individual. Additionally, different classic classifiers have been used for experimentation in order to evaluate the robustness of the proposal. Results confirm the high accuracy of the proposal on the early recognition of people behaviours.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Human behaviour recognition has been, and still remains, a challenging problem that involves different areas of computational intelligence. The automated understanding of people activities from video sequences is an open research topic in which the computer vision and pattern recognition areas have made big efforts. In this paper, the problem is studied from a prediction point of view. We propose a novel method able to early detect behaviour using a small portion of the input, in addition to the capabilities of it to predict behaviour from new inputs. Specifically, we propose a predictive method based on a simple representation of trajectories of a person in the scene which allows a high level understanding of the global human behaviour. The representation of the trajectory is used as a descriptor of the activity of the individual. The descriptors are used as a cue of a classification stage for pattern recognition purposes. Classifiers are trained using the trajectory representation of the complete sequence. However, partial sequences are processed to evaluate the early prediction capabilities having a specific observation time of the scene. The experiments have been carried out using the three different dataset of the CAVIAR database taken into account the behaviour of an individual. Additionally, different classic classifiers have been used for experimentation in order to evaluate the robustness of the proposal. Results confirm the high accuracy of the proposal on the early recognition of people behaviours. |
Azorin-Lopez, Jorge; Fuster-Guillo, Andres; Saval-Calvo, Marcelo; Bradley, David Home Technologies, Smart Systems and eHealth Parte de obra colectiva En: Mechatronic Futures, pp. 179–200, Springer International Publishing, Cham, 2016. @incollection{Azorin-Lopez2016a,
title = {Home Technologies, Smart Systems and eHealth},
author = {Jorge Azorin-Lopez and Andres Fuster-Guillo and Marcelo Saval-Calvo and David Bradley},
url = {http://link.springer.com/10.1007/978-3-319-32156-1_12},
doi = {10.1007/978-3-319-32156-1_12},
year = {2016},
date = {2016-01-01},
booktitle = {Mechatronic Futures},
pages = {179--200},
publisher = {Springer International Publishing},
address = {Cham},
abstract = {Initially, building automation services were provided for larger buildings by means of a set of non-integrated subsystems. By the late 20th century, the concept had developed to include home automation through concepts such as the Digital Home, eHome or iHome based on smart systems and which supported the evolution of the traditional automation services to include entertainment and communication supported by home networks and residential gateways. The 21st century brought with it new paradigms such as “ambient intelligence” and “ubiquitous computing”. In the case of ambient intelligence, this sets out to define a context in which people will be surrounded by intelligent and intuitive interfaces embedded in everyday objects which recognise and respond to their presence and which autonomously and intelligently adapt and respond to their needs while ubiquitous computing distributes processing power throughout the network of smart objects. The evolving constructs now encompass not only the home, but wider concepts such as those of the Smart City. The implementation of smart home and related technologies involves a number of systems issues for the short, medium and long terms such as the choice and future proofing of technologies, ethics and costs. The chapter therefore considers the development and application of smart home technologies and systems before looking at one particular area of application, that of eHealth and mHealth, in more detail.},
keywords = {},
pubstate = {published},
tppubtype = {incollection}
}
Initially, building automation services were provided for larger buildings by means of a set of non-integrated subsystems. By the late 20th century, the concept had developed to include home automation through concepts such as the Digital Home, eHome or iHome based on smart systems and which supported the evolution of the traditional automation services to include entertainment and communication supported by home networks and residential gateways. The 21st century brought with it new paradigms such as “ambient intelligence” and “ubiquitous computing”. In the case of ambient intelligence, this sets out to define a context in which people will be surrounded by intelligent and intuitive interfaces embedded in everyday objects which recognise and respond to their presence and which autonomously and intelligently adapt and respond to their needs while ubiquitous computing distributes processing power throughout the network of smart objects. The evolving constructs now encompass not only the home, but wider concepts such as those of the Smart City. The implementation of smart home and related technologies involves a number of systems issues for the short, medium and long terms such as the choice and future proofing of technologies, ethics and costs. The chapter therefore considers the development and application of smart home technologies and systems before looking at one particular area of application, that of eHealth and mHealth, in more detail. |
2015
|
Saval-Calvo, Marcelo; Azorin-Lopez, Jorge; Fuster-Guillo, Andrés; Mora-Mora, Higinio $mu$-MAR: Multiplane 3D Marker based Registration for depth-sensing cameras Artículo de revista En: Expert Systems with Applications, vol. 42, no. 23, pp. 9353–9365, 2015, ISSN: 09574174. @article{Saval-Calvo2015,
title = {$mu$-MAR: Multiplane 3D Marker based Registration for depth-sensing cameras},
author = {Marcelo Saval-Calvo and Jorge Azorin-Lopez and Andrés Fuster-Guillo and Higinio Mora-Mora},
url = {http://linkinghub.elsevier.com/retrieve/pii/S0957417415005503},
doi = {10.1016/j.eswa.2015.08.011},
issn = {09574174},
year = {2015},
date = {2015-12-01},
journal = {Expert Systems with Applications},
volume = {42},
number = {23},
pages = {9353--9365},
abstract = {Many applications including object reconstruction, robot guidance, and. scene mapping require the registration of multiple views from a scene to generate a complete geometric and appearance model of it. In real situations, transformations between views are unknown and it is necessary to apply expert inference to estimate them. In the last few years, the emergence of low-cost depth-sensing cameras has strengthened the research on this topic, motivating a plethora of new applications. Although they have enough resolution and accuracy for many applications, some situations may not be solved with general state-of-the-art registration methods due to the signal-to-noise ratio (SNR) and the resolution of the data provided. The problem of working with low SNR data, in general terms, may appear in any 3D system, then it is necessary to propose novel solutions in this aspect. In this paper, we propose a method, $mu$-MAR, able to both coarse and fine register sets of 3D points provided by low-cost depth-sensing cameras, despite it is not restricted to these sensors, into a common coordinate system. The method is able to overcome the noisy data problem by means of using a model-based solution of multiplane registration. Specifically, it iteratively registers 3D markers composed by multiple planes extracted from points of multiple views of the scene. As the markers and the object of interest are static in the scenario, the transformations obtained for the markers are applied to the object in order to reconstruct it. Experiments have been performed using synthetic and real data. The synthetic data allows a qualitative and quantitative evaluation by means of visual inspection and Hausdorff distance respectively. The real data experiments show the performance of the proposal using data acquired by a Primesense Carmine RGB-D sensor. The method has been compared to several state-of-the-art methods. The results show the good performance of the $mu$-MAR to register objects with high accuracy in presence of noisy data outperforming the existing methods.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Many applications including object reconstruction, robot guidance, and. scene mapping require the registration of multiple views from a scene to generate a complete geometric and appearance model of it. In real situations, transformations between views are unknown and it is necessary to apply expert inference to estimate them. In the last few years, the emergence of low-cost depth-sensing cameras has strengthened the research on this topic, motivating a plethora of new applications. Although they have enough resolution and accuracy for many applications, some situations may not be solved with general state-of-the-art registration methods due to the signal-to-noise ratio (SNR) and the resolution of the data provided. The problem of working with low SNR data, in general terms, may appear in any 3D system, then it is necessary to propose novel solutions in this aspect. In this paper, we propose a method, $mu$-MAR, able to both coarse and fine register sets of 3D points provided by low-cost depth-sensing cameras, despite it is not restricted to these sensors, into a common coordinate system. The method is able to overcome the noisy data problem by means of using a model-based solution of multiplane registration. Specifically, it iteratively registers 3D markers composed by multiple planes extracted from points of multiple views of the scene. As the markers and the object of interest are static in the scenario, the transformations obtained for the markers are applied to the object in order to reconstruct it. Experiments have been performed using synthetic and real data. The synthetic data allows a qualitative and quantitative evaluation by means of visual inspection and Hausdorff distance respectively. The real data experiments show the performance of the proposal using data acquired by a Primesense Carmine RGB-D sensor. The method has been compared to several state-of-the-art methods. The results show the good performance of the $mu$-MAR to register objects with high accuracy in presence of noisy data outperforming the existing methods. |
Saval-Calvo, Marcelo; Azorin-Lopez, Jorge; Fuster-Guillo, Andres; Garcia-Rodriguez, Jose Three-dimensional planar model estimation using multi-constraint knowledge based on k-means and RANSAC Artículo de revista En: Applied Soft Computing, vol. 34, pp. 572–586, 2015, ISSN: 15684946. @article{Saval-Calvo2015d,
title = {Three-dimensional planar model estimation using multi-constraint knowledge based on k-means and RANSAC},
author = {Marcelo Saval-Calvo and Jorge Azorin-Lopez and Andres Fuster-Guillo and Jose Garcia-Rodriguez},
url = {http://linkinghub.elsevier.com/retrieve/pii/S1568494615003075 https://linkinghub.elsevier.com/retrieve/pii/S1568494615003075},
doi = {10.1016/j.asoc.2015.05.007},
issn = {15684946},
year = {2015},
date = {2015-09-01},
journal = {Applied Soft Computing},
volume = {34},
pages = {572--586},
abstract = {Plane model extraction from three-dimensional point clouds is a necessary step in many different applications such as planar object reconstruction, indoor mapping and indoor localization. Different RANdom SAmple Consensus (RANSAC)-based methods have been proposed for this purpose in recent years. In this study, we propose a novel method-based on RANSAC called Multiplane Model Estimation, which can estimate multiple plane models simultaneously from a noisy point cloud using the knowledge extracted from a scene (or an object) in order to reconstruct it accurately. This method comprises two steps: first, it clusters the data into planar faces that preserve some constraints defined by knowledge related to the object (e.g., the angles between faces); and second, the models of the planes are estimated based on these data using a novel multi-constraint RANSAC. We performed experiments in the clustering and RANSAC stages, which showed that the proposed method performed better than state-of-the-art methods.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Plane model extraction from three-dimensional point clouds is a necessary step in many different applications such as planar object reconstruction, indoor mapping and indoor localization. Different RANdom SAmple Consensus (RANSAC)-based methods have been proposed for this purpose in recent years. In this study, we propose a novel method-based on RANSAC called Multiplane Model Estimation, which can estimate multiple plane models simultaneously from a noisy point cloud using the knowledge extracted from a scene (or an object) in order to reconstruct it accurately. This method comprises two steps: first, it clusters the data into planar faces that preserve some constraints defined by knowledge related to the object (e.g., the angles between faces); and second, the models of the planes are estimated based on these data using a novel multi-constraint RANSAC. We performed experiments in the clustering and RANSAC stages, which showed that the proposed method performed better than state-of-the-art methods. |
Saval-Calvo, Marcelo; Orts-Escolano, Sergio; Azorin-Lopez, Jorge; Garcia-Rodriguez, Jose; Fuster-Guillo, Andres; Morell-Gimenez, Vicente; Cazorla, Miguel Non-rigid point set registration using color and data downsampling Artículo en actas En: 2015 International Joint Conference on Neural Networks (IJCNN), pp. 1–8, IEEE, 2015, ISBN: 978-1-4799-1960-4. @inproceedings{Saval-Calvo2015b,
title = {Non-rigid point set registration using color and data downsampling},
author = {Marcelo Saval-Calvo and Sergio Orts-Escolano and Jorge Azorin-Lopez and Jose Garcia-Rodriguez and Andres Fuster-Guillo and Vicente Morell-Gimenez and Miguel Cazorla},
url = {http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=7280765 http://ieeexplore.ieee.org/document/7280765/},
doi = {10.1109/IJCNN.2015.7280765},
isbn = {978-1-4799-1960-4},
year = {2015},
date = {2015-07-01},
booktitle = {2015 International Joint Conference on Neural Networks (IJCNN)},
pages = {1--8},
publisher = {IEEE},
abstract = {Nowadays, non-rigid registration problem is an active research topic in computer vision. Various proposals exist which face the problem from different perspectives, but it is still a challenging problem. Currently, with the new low-cost RGB-D sensors, the use of both, color and 3D information, is getting more interest in many applications. In this paper, we present a non-rigid registration technique based on CPD, and including color information along with 3D data, to estimate the non-rigid transformation. As the input data size is critical in the processing time, a sampling technique is required. Five sampling techniques are evaluated: a bilinear sampling, a normal-based, a color-based, a combination of the normal and color-based samplings, and a Growing Neural Gas based approach. All of them have been evaluated with the already presented non-rigid registration methods. Results show the performance of each sampling method, obtaining better results for the registration process using color-based sampling techniques.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Nowadays, non-rigid registration problem is an active research topic in computer vision. Various proposals exist which face the problem from different perspectives, but it is still a challenging problem. Currently, with the new low-cost RGB-D sensors, the use of both, color and 3D information, is getting more interest in many applications. In this paper, we present a non-rigid registration technique based on CPD, and including color information along with 3D data, to estimate the non-rigid transformation. As the input data size is critical in the processing time, a sampling technique is required. Five sampling techniques are evaluated: a bilinear sampling, a normal-based, a color-based, a combination of the normal and color-based samplings, and a Growing Neural Gas based approach. All of them have been evaluated with the already presented non-rigid registration methods. Results show the performance of each sampling method, obtaining better results for the registration process using color-based sampling techniques. |
Azorin-Lopez, Jorge; Saval-Calvo, Marcelo; Fuster-Guillo, Andres; Garcia-Rodriguez, Jose; Orts-Escolano, Sergio Self-Organizing Activity Description Map to represent and classify human behaviour Artículo en actas En: 2015 International Joint Conference on Neural Networks (IJCNN), pp. 1–7, IEEE, 2015, ISBN: 978-1-4799-1960-4. @inproceedings{Azorin-Lopez2015c,
title = {Self-Organizing Activity Description Map to represent and classify human behaviour},
author = {Jorge Azorin-Lopez and Marcelo Saval-Calvo and Andres Fuster-Guillo and Jose Garcia-Rodriguez and Sergio Orts-Escolano},
url = {http://ieeexplore.ieee.org/document/7280784/},
doi = {10.1109/IJCNN.2015.7280784},
isbn = {978-1-4799-1960-4},
year = {2015},
date = {2015-07-01},
booktitle = {2015 International Joint Conference on Neural Networks (IJCNN)},
pages = {1--7},
publisher = {IEEE},
abstract = {The automated understanding of people activities from video sequences is an open research topic in which the computer vision and pattern recognition areas have made big efforts in recent years. This paper proposes the Self Organizing Activity Description Map (SOADM). It is a novel neural network based on the self-organizing paradigm to classify high level of semantic understanding from video sequences. The neural network is able to deal with the big gap between human trajectories in a scene and the global behaviour associated to them. Specifically, using simple representations of people trajectories as input, the SOADM is able to both represent and classify human behaviours. Additionally, the map is able to preserve the topological information about the scene. Experiments have been carried out using the Shopping Centre dataset of the CAVIAR database taken into account the global behaviour of an individual. Results confirm the high accuracy of the proposal outperforming previous methods.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The automated understanding of people activities from video sequences is an open research topic in which the computer vision and pattern recognition areas have made big efforts in recent years. This paper proposes the Self Organizing Activity Description Map (SOADM). It is a novel neural network based on the self-organizing paradigm to classify high level of semantic understanding from video sequences. The neural network is able to deal with the big gap between human trajectories in a scene and the global behaviour associated to them. Specifically, using simple representations of people trajectories as input, the SOADM is able to both represent and classify human behaviours. Additionally, the map is able to preserve the topological information about the scene. Experiments have been carried out using the Shopping Centre dataset of the CAVIAR database taken into account the global behaviour of an individual. Results confirm the high accuracy of the proposal outperforming previous methods. |
Orts-Escolano, Sergio; Garcia-Rodriguez, Jose; Morell, Vicente; Cazorla, Miguel; Saval, Marcelo; Azorin, Jorge Processing point cloud sequences with Growing Neural Gas Artículo en actas En: 2015 International Joint Conference on Neural Networks (IJCNN), pp. 1–8, IEEE, 2015, ISBN: 978-1-4799-1960-4. @inproceedings{Orts-Escolano2015,
title = {Processing point cloud sequences with Growing Neural Gas},
author = {Sergio Orts-Escolano and Jose Garcia-Rodriguez and Vicente Morell and Miguel Cazorla and Marcelo Saval and Jorge Azorin},
url = {http://ieeexplore.ieee.org/document/7280709/},
doi = {10.1109/IJCNN.2015.7280709},
isbn = {978-1-4799-1960-4},
year = {2015},
date = {2015-07-01},
booktitle = {2015 International Joint Conference on Neural Networks (IJCNN)},
pages = {1--8},
publisher = {IEEE},
abstract = {We consider the problem of processing point cloud sequences. In particular, we represent and track objects in dynamic scenes acquired using low-cost sensors such as the Kinect. A neural network based approach is proposed to represent and estimate 3D objects motion. This system addresses multiple computer vision tasks such as object segmentation, representation, motion analysis and tracking. The use of a neural network allows the unsupervised estimation of motion and the representation of objects in the scene. This proposal avoids the problem of finding corresponding features while tracking moving objects. A set of experiments are presented that demonstrate the validity of our method to track 3D objects. Favorable results are presented demonstrating the capabilities of the GNG algorithm for this task.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
We consider the problem of processing point cloud sequences. In particular, we represent and track objects in dynamic scenes acquired using low-cost sensors such as the Kinect. A neural network based approach is proposed to represent and estimate 3D objects motion. This system addresses multiple computer vision tasks such as object segmentation, representation, motion analysis and tracking. The use of a neural network allows the unsupervised estimation of motion and the representation of objects in the scene. This proposal avoids the problem of finding corresponding features while tracking moving objects. A set of experiments are presented that demonstrate the validity of our method to track 3D objects. Favorable results are presented demonstrating the capabilities of the GNG algorithm for this task. |
Azorin-Lopez, J.; Garcia-Rodriguez, J; Jimeno-Morenilla, A; Mora-Mora, H; Pujol-Lopez, F; Sanchez-Romero, J L; Saval-Calvo, M; Orts-Escolano, S; Morell-Gimenez, V; Garcia-Garcia, A; Rizo-Gomez, A AN EXPERIENCE OF SPECIFIC LEARNING METHODOLOGIES IN COMPUTER ARCHITECTURE FOR GROUPS OF HIGH ACADEMIC PERFORMANCE Artículo en actas En: 9th International Technology, Education and Development Conference, pp. 3591–3601, 2015. @inproceedings{J.Azorin-Lopez2015,
title = {AN EXPERIENCE OF SPECIFIC LEARNING METHODOLOGIES IN COMPUTER ARCHITECTURE FOR GROUPS OF HIGH ACADEMIC PERFORMANCE},
author = {J. Azorin-Lopez and J Garcia-Rodriguez and A Jimeno-Morenilla and H Mora-Mora and F Pujol-Lopez and J L Sanchez-Romero and M Saval-Calvo and S Orts-Escolano and V Morell-Gimenez and A Garcia-Garcia and A Rizo-Gomez},
year = {2015},
date = {2015-01-01},
booktitle = {9th International Technology, Education and Development Conference},
pages = {3591--3601},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
|
Saval-Calvo, Marcelo; Orts-Escolano, Sergio; Azorin-Lopez, Jorge; Garcia-Rodriguez, Jose; Fuster-Guillo, Andres; Morell-Gimenez, Vicente; Cazorla, Miguel A Comparative Study of Downsampling Techniques for Non-rigid Point Set Registration Using Color Parte de obra colectiva En: 6th International Work-Conference on the Interplay between Natural and Artificial Computation, 2015, pp. 281–290, 2015. @incollection{Saval-Calvo2015a,
title = {A Comparative Study of Downsampling Techniques for Non-rigid Point Set Registration Using Color},
author = {Marcelo Saval-Calvo and Sergio Orts-Escolano and Jorge Azorin-Lopez and Jose Garcia-Rodriguez and Andres Fuster-Guillo and Vicente Morell-Gimenez and Miguel Cazorla},
url = {http://link.springer.com/10.1007/978-3-319-18833-1_30},
doi = {10.1007/978-3-319-18833-1_30},
year = {2015},
date = {2015-01-01},
booktitle = {6th International Work-Conference on the Interplay between Natural and Artificial Computation, 2015},
pages = {281--290},
abstract = {Registration of multiple sets of data into a common coordinate system is an important problem in many areas of computer vision and robotics. Usually a large set of data is involved in the process. Moreover, the sets are in general composed by a large number of 3D points. The input for registration techniques based on point set as inputs make sometimes intractable the process due to time needed to provide a feasible solution to the transformation between data. This problem is harder when the transformation is non-rigid. Correspondence estimation and transformation is usually done for each point in the data set. The size of the input is critical for the processing time and, in consequence, a sampling technique is previously required. In this paper, a comparative study of five sampling techniques is carried out. Specifically, is considered a bilinear sampling, a normal-based, a color-based, a combination of the normal and color-based samplings, and a Growing Neural Gas (GNG) based approach. They have been evaluated to reduce the number of points in the input of two non-rigid registration techniques: the Coherent Point Drift (CPD) and our proposal of a non-rigid registration technique based on CPD that includes color information.},
keywords = {},
pubstate = {published},
tppubtype = {incollection}
}
Registration of multiple sets of data into a common coordinate system is an important problem in many areas of computer vision and robotics. Usually a large set of data is involved in the process. Moreover, the sets are in general composed by a large number of 3D points. The input for registration techniques based on point set as inputs make sometimes intractable the process due to time needed to provide a feasible solution to the transformation between data. This problem is harder when the transformation is non-rigid. Correspondence estimation and transformation is usually done for each point in the data set. The size of the input is critical for the processing time and, in consequence, a sampling technique is previously required. In this paper, a comparative study of five sampling techniques is carried out. Specifically, is considered a bilinear sampling, a normal-based, a color-based, a combination of the normal and color-based samplings, and a Growing Neural Gas (GNG) based approach. They have been evaluated to reduce the number of points in the input of two non-rigid registration techniques: the Coherent Point Drift (CPD) and our proposal of a non-rigid registration technique based on CPD that includes color information. |
Azorin-Lopez, Jorge; Saval-Calvo, Marcelo; Fuster-Guillo, Andres; Mora-Mora, Higinio; Villena-Martinez, Victor Topology Preserving Self-Organizing Map of Features in Image Space for Trajectory Classification Parte de obra colectiva En: 6th International Work-Conference on the Interplay between Natural and Artificial Computation, 2015, pp. 271–280, 2015. @incollection{Azorin-Lopez2015b,
title = {Topology Preserving Self-Organizing Map of Features in Image Space for Trajectory Classification},
author = {Jorge Azorin-Lopez and Marcelo Saval-Calvo and Andres Fuster-Guillo and Higinio Mora-Mora and Victor Villena-Martinez},
url = {http://link.springer.com/10.1007/978-3-319-18833-1_29},
doi = {10.1007/978-3-319-18833-1_29},
year = {2015},
date = {2015-01-01},
booktitle = {6th International Work-Conference on the Interplay between Natural and Artificial Computation, 2015},
pages = {271--280},
abstract = {Self-Organizing maps (SOM) are able to preserve topological information in the projecting space. Structure and learning algorithm of SOMs restrict the topological preservation in the map. Adjacent neurons share similar vector features. However, topological preservation from the input space is not always accomplished. In this paper, we propose a novel self-organizing feature map that is able to preserve the topological information about the scene in the image space. Extracted features in adjacent areas of an image are explicitly in adjacent areas of the self-organizing map preserving input topology (SOM-PINT). The SOM-PINT has been applied to represent and classify trajectories into high level of semantic understanding from video sequences. Experiments have been carried out using the Shopping Centre dataset of the CAVIAR database taken into account the global behaviour of an individual. Results confirm the input preservation topology in image space to obtain high performance classification for trajectory classification in contrast of traditional SOM.},
keywords = {},
pubstate = {published},
tppubtype = {incollection}
}
Self-Organizing maps (SOM) are able to preserve topological information in the projecting space. Structure and learning algorithm of SOMs restrict the topological preservation in the map. Adjacent neurons share similar vector features. However, topological preservation from the input space is not always accomplished. In this paper, we propose a novel self-organizing feature map that is able to preserve the topological information about the scene in the image space. Extracted features in adjacent areas of an image are explicitly in adjacent areas of the self-organizing map preserving input topology (SOM-PINT). The SOM-PINT has been applied to represent and classify trajectories into high level of semantic understanding from video sequences. Experiments have been carried out using the Shopping Centre dataset of the CAVIAR database taken into account the global behaviour of an individual. Results confirm the input preservation topology in image space to obtain high performance classification for trajectory classification in contrast of traditional SOM. |
2014
|
Morell-Gimenez, Vicente; Saval-Calvo, Marcelo; Azorin-Lopez, Jorge; Garcia-Rodriguez, Jose; Cazorla, Miguel; Orts-Escolano, Sergio; Fuster-Guillo, Andres A Comparative Study of Registration Methods for RGB-D Video of Static Scenes Artículo de revista En: Sensors, vol. 14, no. 5, pp. 8547–8576, 2014, ISSN: 1424-8220. @article{Morell-Gimenez2014,
title = {A Comparative Study of Registration Methods for RGB-D Video of Static Scenes},
author = {Vicente Morell-Gimenez and Marcelo Saval-Calvo and Jorge Azorin-Lopez and Jose Garcia-Rodriguez and Miguel Cazorla and Sergio Orts-Escolano and Andres Fuster-Guillo},
url = {http://www.mdpi.com/1424-8220/14/5/8547/ http://www.mdpi.com/1424-8220/14/5/8547},
doi = {10.3390/s140508547},
issn = {1424-8220},
year = {2014},
date = {2014-05-01},
journal = {Sensors},
volume = {14},
number = {5},
pages = {8547--8576},
abstract = {The use of RGB-D sensors for mapping and recognition tasks in robotics or, in general, for virtual reconstruction has increased in recent years. The key aspect of these kinds of sensors is that they provide both depth and color information using the same device. In this paper, we present a comparative analysis of the most important methods used in the literature for the registration of subsequent RGB-D video frames in static scenarios. The analysis begins by explaining the characteristics of the registration problem, dividing it into two representative applications: scene modeling and object reconstruction. Then, a detailed experimentation is carried out to determine the behavior of the different methods depending on the application. For both applications, we used standard datasets and a new one built for object reconstruction.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The use of RGB-D sensors for mapping and recognition tasks in robotics or, in general, for virtual reconstruction has increased in recent years. The key aspect of these kinds of sensors is that they provide both depth and color information using the same device. In this paper, we present a comparative analysis of the most important methods used in the literature for the registration of subsequent RGB-D video frames in static scenarios. The analysis begins by explaining the characteristics of the registration problem, dividing it into two representative applications: scene modeling and object reconstruction. Then, a detailed experimentation is carried out to determine the behavior of the different methods depending on the application. For both applications, we used standard datasets and a new one built for object reconstruction. |
Azorín-López, Jorge; Saval-Calvo, Marcelo; Fuster-Guilló, Andrés; Oliver-Albert, Antonio A predictive model for recognizing human behaviour based on trajectory representation Artículo en actas En: Neural Networks (IJCNN), 2014 International Joint Conference on, pp. 1494–1501, 2014. @inproceedings{Azorin-Lopez2014,
title = {A predictive model for recognizing human behaviour based on trajectory representation},
author = {Jorge Azorín-López and Marcelo Saval-Calvo and Andrés Fuster-Guilló and Antonio Oliver-Albert},
doi = {10.1109/IJCNN.2014.6889883},
year = {2014},
date = {2014-01-01},
booktitle = {Neural Networks (IJCNN), 2014 International Joint Conference on},
pages = {1494--1501},
abstract = {The automatic understanding of the behaviour conducted by humans in scenarios using images as input of the system is a very important and challenging problem involving different areas of computational intelligence. In this paper human activity recognition is studied from a prediction point of view. We propose a model that, in addition to the capabilities of it to predict behaviour from new inputs, it is able to detect behaviour using a portion of the input. Specifically, we propose a prediction activity method based on the Activity Description Vector (ADV) to early detect the behaviour performed by a person in a scene. ADV is used to extract features that are normalized to be the cue of behaviour classifiers. We use complete sequences for training and partial sequences to evaluate the prediction capabilities having a specific observation time of the scene. CAVIAR dataset and different classic classifiers have been used for experimentation in order to evaluate the proposal obtaining great accuracy on the early recognition.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The automatic understanding of the behaviour conducted by humans in scenarios using images as input of the system is a very important and challenging problem involving different areas of computational intelligence. In this paper human activity recognition is studied from a prediction point of view. We propose a model that, in addition to the capabilities of it to predict behaviour from new inputs, it is able to detect behaviour using a portion of the input. Specifically, we propose a prediction activity method based on the Activity Description Vector (ADV) to early detect the behaviour performed by a person in a scene. ADV is used to extract features that are normalized to be the cue of behaviour classifiers. We use complete sequences for training and partial sequences to evaluate the prediction capabilities having a specific observation time of the scene. CAVIAR dataset and different classic classifiers have been used for experimentation in order to evaluate the proposal obtaining great accuracy on the early recognition. |
2013
|
Azorin-Lopez, Jorge; Saval-Calvo, Marcelo; Fuster-Guillo, Andres; Garcia-Rodriguez, Jose Human behaviour recognition based on trajectory analysis using neural networks Artículo en actas En: The 2013 International Joint Conference on Neural Networks (IJCNN), pp. 1–7, IEEE, 2013, ISBN: 978-1-4673-6129-3. @inproceedings{Azorin-Lopez2013,
title = {Human behaviour recognition based on trajectory analysis using neural networks},
author = {Jorge Azorin-Lopez and Marcelo Saval-Calvo and Andres Fuster-Guillo and Jose Garcia-Rodriguez},
url = {http://ieeexplore.ieee.org/document/6706724/},
doi = {10.1109/IJCNN.2013.6706724},
isbn = {978-1-4673-6129-3},
year = {2013},
date = {2013-08-01},
booktitle = {The 2013 International Joint Conference on Neural Networks (IJCNN)},
pages = {1--7},
publisher = {IEEE},
abstract = {Automated human behaviour analysis has been, and still remains, a challenging problem. It has been dealt from different points of views: from primitive actions to human interaction recognition. This paper is focused on trajectory analysis which allows a simple high level understanding of complex human behaviour. It is proposed a novel representation method of trajectory data, called Activity Description Vector (ADV) based on the number of occurrences of a person is in a specific point of the scenario and the local movements that perform in it. The ADV is calculated for each cell of the scenario in which it is spatially sampled obtaining a cue for different clustering methods. The ADV representation has been tested as the input of several classic classifiers and compared to other approaches using CAVIAR dataset sequences obtaining great accuracy in the recognition of the behaviour of people in a Shopping Centre.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Automated human behaviour analysis has been, and still remains, a challenging problem. It has been dealt from different points of views: from primitive actions to human interaction recognition. This paper is focused on trajectory analysis which allows a simple high level understanding of complex human behaviour. It is proposed a novel representation method of trajectory data, called Activity Description Vector (ADV) based on the number of occurrences of a person is in a specific point of the scenario and the local movements that perform in it. The ADV is calculated for each cell of the scenario in which it is spatially sampled obtaining a cue for different clustering methods. The ADV representation has been tested as the input of several classic classifiers and compared to other approaches using CAVIAR dataset sequences obtaining great accuracy in the recognition of the behaviour of people in a Shopping Centre. |
Saval-Calvo, Marcelo; Azorín-López, Jorge; Fuster-Guilló, Andrés Model-Based Multi-view Registration for RGB-D Sensors Parte de obra colectiva En: 12th International Work-Conference on Artificial Neural Networks, IWANN 2013,, pp. 496–503, 2013. @incollection{Saval-Calvo2013,
title = {Model-Based Multi-view Registration for RGB-D Sensors},
author = {Marcelo Saval-Calvo and Jorge Azorín-López and Andrés Fuster-Guilló},
url = {http://link.springer.com/10.1007/978-3-642-38682-4_53},
doi = {10.1007/978-3-642-38682-4_53},
year = {2013},
date = {2013-01-01},
booktitle = {12th International Work-Conference on Artificial Neural Networks, IWANN 2013,},
pages = {496--503},
abstract = {Registration is a main task in 3D objects reconstruction. Different approaches have been developed in order to solve specific problems in scenarios, objects or even the source of data. Recently, new problems have been appeared with the increasing use of low-cost RGB-D sensors. Registering small objects acquired by these cameras using traditional methods is a hard problem due to their low resolution and depth sensing error. In this paper, we propose a model-based registration method for objects composed by small planes using multi-view acquisition. It is able to deal with the problem of low resolution and noisy data. Experiments show very good promising results registering small objects acquired with low-cost RGB-D sensors compared to ICP variants.},
keywords = {},
pubstate = {published},
tppubtype = {incollection}
}
Registration is a main task in 3D objects reconstruction. Different approaches have been developed in order to solve specific problems in scenarios, objects or even the source of data. Recently, new problems have been appeared with the increasing use of low-cost RGB-D sensors. Registering small objects acquired by these cameras using traditional methods is a hard problem due to their low resolution and depth sensing error. In this paper, we propose a model-based registration method for objects composed by small planes using multi-view acquisition. It is able to deal with the problem of low resolution and noisy data. Experiments show very good promising results registering small objects acquired with low-cost RGB-D sensors compared to ICP variants. |
2012
|
Saval-Calvo, Marcelo; Azorín-López, Jorge; Fuster-Guilló, Andrés Comparative Analysis of Temporal Segmentation Methods of Video Sequences Parte de obra colectiva En: Garcia-Rodriguez, Jose; Quevedo, Miguel A Cazorla (Ed.): Robotic Vision, IGI Global, 2012, ISBN: 9781466626720. @incollection{Saval-Calvo2012a,
title = {Comparative Analysis of Temporal Segmentation Methods of Video Sequences},
author = {Marcelo Saval-Calvo and Jorge Azorín-López and Andrés Fuster-Guilló},
editor = {Jose Garcia-Rodriguez and Miguel A Cazorla Quevedo},
url = {http://www.igi-global.com/chapter/comparative-analysis-temporal-segmentation-methods/73183/},
doi = {10.4018/978-1-4666-2672-0},
isbn = {9781466626720},
year = {2012},
date = {2012-07-01},
booktitle = {Robotic Vision},
publisher = {IGI Global},
abstract = {In this chapter, a comparative analysis of basic segmentation methods of video sequences and their combinations is carried out. Analysis of different algorithms is based on the efficiency (true positive and false positive rates) and temporal cost to provide regions in the scene. These are two of the most important requirements of the design to provide to the tracking with segmentation in an efficient and timely manner constrained to the application. Specifically, methods using temporal information as Background Subtraction, Temporal Differencing, Optical Flow, and the four combinations of them have been analyzed. Experimentation has been done using image sequences of CAVIAR project database. Efficiency results show that Background Subtraction achieves the best individual result whereas the combination of the three basic methods is the best result in general. However, combinations with Optical Flow should be considered depending of application, because its temporal cost is too high with respect to efficiency provided to the combination.},
keywords = {},
pubstate = {published},
tppubtype = {incollection}
}
In this chapter, a comparative analysis of basic segmentation methods of video sequences and their combinations is carried out. Analysis of different algorithms is based on the efficiency (true positive and false positive rates) and temporal cost to provide regions in the scene. These are two of the most important requirements of the design to provide to the tracking with segmentation in an efficient and timely manner constrained to the application. Specifically, methods using temporal information as Background Subtraction, Temporal Differencing, Optical Flow, and the four combinations of them have been analyzed. Experimentation has been done using image sequences of CAVIAR project database. Efficiency results show that Background Subtraction achieves the best individual result whereas the combination of the three basic methods is the best result in general. However, combinations with Optical Flow should be considered depending of application, because its temporal cost is too high with respect to efficiency provided to the combination. |
