Research

Research interests

My research activities are developed at the Computer Vision Center, as part of the Barcelona Perceptual Computing Lab, and at the Internet Interdisciplinary Institute as part of the Scene Understanding and Artificial Intelligence Lab. My research interests are related to machine learning, evolutionary computation, and statistical pattern recognition. Recently, we are opening a new research line to develop algorithms for generic object recognition over huge cardinality image databases.

Projects

Traffic Sign Recognition: This is a project started at 2004 in collaboration with the Institut Cartogràfic de Catalunya, where the images captured in the context of a mobile mapping project are analyzed in order to extract useful information. There are basically two main considered scenarios: road and urban scenes. In each scenario the objects of interest and the difficulties are different. During this project different detection and classification techniques have been developed in order to deal with traffic signs recognition and text location.

 

BCNMining:  The goal of this project the goal is to develop new methods to index the visual information contained in a large image database. This database has been generated by the same mobile mapping system that in the traffic sign recognition project. In this case, only urban images from the city of Barcelona have been considered. Our primary goal is to find representations and indexing methodologies to perform image searches of any type of objects.

 

Evolutionary ECOC: In this line, we are working on sub-linear Error Correcting Output Codes (ECOC). The goal is to find effective and reliable solutions for the multi-class classification problem, in special attention to the applicability to large databases with a huge number of classes. We are working on the application of evolutionary algorithms in order to face with the large dimensionality and cardinality of new problems.

 

Computer Vision in Museums: This project attempts improve the Museums visitors experience using new applications based on Computer Vision Techniques and Augmented Reality.

 

Human Pose Recovery and Behaviour Analysis: This is the current main research line. Our goal is to analyse scenes in a human centric view. Some of the topics we are interested in are: Human detection and segmentation, pose estimation, action recognition, social signal, user recognition, gesture recognition, facial expression analysis and in general any behavioural aspects of humans in a given scene and context.

 

Publications

2015

David Bañeras, Xavier Baró, Ana Elena Guerrero, Laura Porta, Elena Rodríguez. Experiencias en la evaluación de estudiantes residentes en el extranjero. XXI Jornadas sobre la Enseñanza Universitaria de la Informática (JENUI).

2014

Miguel Ángel Bautista, Sergio Escalera, Xavier Baró, and Oriol Pujol. On the design of an ECOC-compliant genetic algorithm. Pattern Recognition 47 (2), pp. 865-884. ISSN 0031-3203. February 2014.
Sergio Escalera, Xavier Baró, Jordi González, Miguel Ángel Bautista, , Meysam Madadi, Miguel Reyes, Victor Ponce, Hugo J. Escalante, Jamie Shotton, and Isabelle Guyon. ChaLearn Looking at People Challenge 2014: Dataset and Results. Lecture Notes in Computer Science, pp. 459-473. ISSN.0302-9743.
Joan Arnedo, David Bañeras, Xavier Baró, Santi Caballé, Ana Elena Guerrero, Xavier Mas, Laura Porta, and Josep Prieto. Val-ID: A trust-based virtual assessment system. Proceedings of the 6-th International Conference on Intelligent Networking and Collaborative Systems (INCoS-2014). Salerno: IEEE Press.

2013

David Sanchez-Mendoza, David Masip, Xavier Baró, and Àgata Lapedriza . Emotion Detection Using Hybrid Structural and Appearance Descriptors. Proceedings of the 10th International Conference, MDAI 2013, Barcelona, Spain, November 20-22, 2013.
Antonio Hernández-Vela, Miguel Ángel Bautista, Xavier Perez-Sala, Víctor Ponce-López, Sergio Escalera, Xavier Baró, Oriol Pujol, and Cecilio Angulo. Probability-based Dynamic Time Warping and Bag-of-Visual-and-Depth-Words for Human Gesture Recognition in RGB-D. Pattern Recognition Letters. ISSN 0167-8655. September 2013.
Víctor Ponce-López, Sergio Escalera, and Xavier Baró. Multi-modal social signal analysis for predicting agreement in conversation settings. Proceedings of the 15th ACM on International conference on multimodal interaction, pp. 495-502. December 2013.
Sergio Escalera, Jordi Gonzàlez, Xavier Baró, Miguel Reyes, Oscar Lopes, Isabelle Guyon, Vassilis Athitsos, and Hugo Escalante. Multi-modal gesture recognition challenge 2013: dataset and results. Proceedings of the 15th ACM on International conference on multimodal interaction, pp. 445-452. December 2013.
Sergio Escalera, Jordi Gonzàlez, Xavier Baró, Miguel Reyes, Isabelle Guyon, Vassilis Athitsos, Hugo Escalante, Leonid Sigal, Antonis Argyros, Cristian Sminchisescu, Richard Bowden, and Stan Sclaroff. ChaLearn multi-modal gesture recognition 2013: grand challenge and workshop summary. Proceedings of the 15th ACM on International conference on multimodal interaction, pp. 365-368. December 2013.
Laura Igual and Xavier Baro Experiencia de aprendizaje de programación basada en proyectos. Simposio-Taller Estrategias y herramientas para el aprendizaje y la evaluación. JENUI 2013. July 2013.
Xavier Baro, David Masip, Elena Planas, and Julia Minguillon. PeLP: Plataforma para el Aprendizaje de Lenguajes de Programación. Recursos Docentes. JENUI 2013. July 2013.

2012

Miguel Ángel Bautista, Antonio Hernández-Vela, Victor Ponce, Xavier Perez-Sala, Xavier Baró, Oriol Pujol, Cecilio Angulo, and Sergio Escalera. Probability-based Dynamic Time Warping for Gesture Recognition on RGB-D data. International Workshop on Depth Image Analysis, ICPR 2012, Tsukuba Science City, Japan.
Antonio Hernández-Vela, Miguel Ángel Bautista, Xavier Perez-Sala, Víctor Ponce, Xavier Baró, Oriol Pujol, Cecilio Angulo, and Sergio Escalera. BoVDW: Bag-of-Visual-and-Depth-Words for Gesture Recognition. 21st International Conference on Pattern Recognition, November 2012, Tsukuba Science City, Japan.
Miguel Angel Bautista, Sergio Escalera, Xavier Baró, and Oriol Pujol. A Genetic Inspired Optimization for ECOC. Joint IAPR International Workshops on Structural and Syntactic Pattern Recognition and Statistical Techniques in Pattern Recognition (ICPR 2012), Lecture Notes in Computer Science Volume 7626, pp 743-751, November 2012.
Sergio Escalera, Xavier Baró, Jordi Vitrià, Petia Radeva and Bogdan Raducanu. Social Network Extraction and Analysis Based on Multimodal Dyadic Interaction. Sensors. 12(2) pp. 1702-1719. ISSN: 1424-8220. January 2012.
Àngels Rius, Xavier Aracíl, Xavier Baró. UOC API Site, a seed for new eLearning applications. Proceedings of the IX Multidisciplinary Symposium on the Design and Evaluation of Digital Content for Education, pp. 245-252. June 2012.

2011

Miguel Ángel Bautista, Sergio Escalera, Xavier Baró, Petia Radeva, Jordi Vitria, and Oriol Pujol Minimal Design of Error-Correcting Output Codes. Pattern Recognition Letters,Volume 33 Issue 6, pp. 693-702 . ISSN: 0167-8655. April 2012.
Sergio Escalera, Xavier Baró, Oriol Pujol, Jordi Vitrià, and Petia Radeva Traffic-Sign Recognition Systems. SpringerBriefs in Computer Science. Springer. ISBN 978-1-4471-2244-9, November 2011.
Víctor Ponce, Mario Gorga, Xavier Baró, Petia Radeva, and Sergio Escalera Análisis de la expresión oral y gestual en proyectos fin de carrera vía un sistema de visión artificial. ReVision, la revista electrónica de la Asociación de Enseñantes Universitarios de la Informática AENUI, from JENUI 2010 best work. Pages 8-18 Num 1, Vol 4. ISSN 1989-1199, June 2011.
Miguel Angel Bautista, Oriol Pujol, Xavier Baró, and Sergio Escalera Introducing Separability Matrix for Error Correcting Output Codes coding. Lecture Notes in Computer Science, 6713. pp. 227-236. ISSN 0302-9743. June 2011
Victor Ponce, Mario Gorga, Xavier Baró, and Sergio Escalera Human Behavior Analysis from Video Data using Bag-of-Gestures.Proceedings of the Twenty-Second International Joint Conference on Artificial Intelligence, pp. 2836-2837. AAAI Press, USA. ISBN: 978-1-57735-516-8
Michal Drozdal, Antoni Hernández, Santiago Seguí, Xavier Baró, Sergio Escalera, Àgata Lapedriza, David Masip, Petia Radeva, and Jordi Vitrià Combining Detectors for Human Layout Analysis.In: Maragos, P., Paragios, N., Daniliidis, K. 11th European Conference on Computer Vision, The Pascal Visual Object Classes Challenge Workshop 2010. Fort North: Pag. 1.

2010

Sergio Escalera, Xavier Baró, Jordi Vitrià, Petia Radeva and Bogdan Raducanu. Modelling and Analyzing Multimodal Dyadic Interactions Using Social Networks.Proceedings of the 12th International Conference on Multimodal Interfaces and 7th Workshop on Machine Learning for Multimodal Interaction (ICMI-MLMI 2010). ISBN: 978-1-4503-0414-6. ACM Press. Beijing, China, November 8-12, 2010.
Miguel A. Bautista, Xavier Baró, Oriol Pujol, Petia Radeva, Jordi Vitrià, and Sergio Escalera. Compact Evolutive Design of Error-Correcting Output Codes. Proceedings of the Supervised and Unsupervised Ensemble Methods and their Applications (SUEMA’2010) in the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML PKDD), pp. 119-128. Barcelona, September 20th to 24th, 2010.
Sergio Escalera, Xavier Baró and Petia Radeva. Métodos automáticos para el análisis de la expresión oral y gestual en proyectos fin de carrera.In JENUI’10: Proceedings of the XVI Jornadas de Enseñanza Universitaria de la Informatica, pp. 67-74. ISBN: 84-693-3741-7. Santiago de Compostela, Spain, July, 2010
Miguel A. Bautista, Sergio Escalera, Xavier Baró, Oriol Pujol, Petia Radeva, and Jordi Vitrià. Compact Design of ECOC for Multi-class Object Categorization.5th CVC Workshop on Research and Development: Achievements and New Opportunities in Computer Vision, Ediciones Gráficas Rey, S.L. ISBN: 978-84-938351-0-1, pp. 54-57, October 2010.
Victor Ponce, Sergio Escalera, Xavier Baró, and Petia Radeva. Automatic Analysis of Non-verbal Communication.5th CVC Workshop on Research and Development: Achievements and New Opportunities in Computer Vision, Ediciones Gráficas Rey, S.L. ISBN: 978-84-938351-0-1, pp. 105-108, October 2010.

2009

Xavier Baró, Sergio Escalera, and Miquel Ferrer (eds.). 4th CVC Workshop on Research and Development: New Trends and Challenges in Computer Vision.Ediciones Gráficas Rey, S.L. ISBN: 978-84-937261-1-9. October 2009.
Sergio Escalera, Xavier Baró, Jordi Vitrià, and Petia Radeva Text Detection in Urban Scenes. Artificial Intelligence Resarch and Development, Frontiers in Artificial Intelligence and Applications, Vol. 202 . pp. 35-44. ISBN: 978-1-60750-061-2. October 2009.
Xavier Baró, Sergio Escalera, Petia Radeva, and Jordi Vitrià Generic Object Recognition in Urban Image Databases. Artificial Intelligence Resarch and Development, Frontiers in Artificial Intelligence and Applications, Vol. 202 . pp. 27-34. ISBN: 978-1-60750-061-2. October 2009.
Xavier Baró and Jordi Vitrià Probabilistic Darwin Machines for Object Detection. In Proceedings of the 2009 International Conference on Image Processing, Computer Vision, and Pattern Recognition. July 2009.
Xavier Baró, Sergio Escalera, Petia Radeva, and Jordi Vitrià Visual Content Layer for Scalable Object Recognition in Urban Image Databases.In Proceedings of the 2009 IEEE International Conference on Multimedia and Expo, pp.1616-1619. ISSN: 1945-7871. July 2009.
Xavier Baró, Sergio Escalera, Jordi Vitrià, Oriol Pujol, and Petia Radeva Traffic Sign Recognition Using Evolutionary Adaboost Detection and Forest-ECOC Classification. IEEE Transactions on Intelligent Transportation Systems, 10 (1). pp. 113-126. ISSN 1024-8072. March 2009.
Xavier Baró Probabilistic Darwin Machines: A new approach to develop Evolutionary Object Detection Systems. Ediciones Gráficas Rey, S.L. ISBN 978-84-936529-2-0. April 2009.

2008

Xavier Baró and Jordi Vitrià Evolutionary Object Detection by means of Naïve Bayes Models Estimation. Lecture Notes in Computer Science, 4974. pp. 235-244. ISSN 0302-9743. March 2008.
Xavier Baró and Jordi Vitrià Weighted Dissociated Dipoles: An Extended Visual Feature Set. Lecture Notes in Computer Science, Vol. 5008. pp. 281-290. ISSN 0302-9743. May 2008.

2007

Xavier Baró and Jordi Vitrià Weighted Dissociated Dipoles for Evolutive Learning. Artificial Intelligence Resarch and Development, Frontiers in Artificial Intelligence and Applications, Vol. 163 . pp. 189-196. ISBN: 978-1-58603-798-7. October 2007.
Xavier Baró and Jordi Vitrià Visual features selection using Naïve Bayes Models. Computer Vision: Process of research and development, pp. 54-59. ISBN: 978-84-935251-4-9. October 2007.

2006

Xavier Baró and Jordi Vitrià Real-time object detection using an evolutionary boosting strategy. Artificial Intelligence Resarch and Development, Frontiers in Artificial Intelligence and Applications, Vol. 146 . pp. 9-18. ISBN: 978-1-58603-663-7. October 2006.
Xavier Baró and Jordi Vitrià Evolutionary boosting strategy. Computer Vision: Process of research and development, pp. 59-62. ISBN: 978-84-933652-8-9. October 2006.

2005

Xavier Baró and Jordi Vitrià Feature Selection with Non-Parametric Mutual Information for Adaboost Learning. Artificial Intelligence Resarch and Development, Frontiers in Artificial Intelligence and Applications, Vol. 131 . pp. 131-138. ISBN: 978-1-58603-560-6. October 2005.

2004

Xavier Baró and Jordi Vitrià Traffic sign detection on grayscale images. Recent advances in artificial intelligence research and development, Frontiers in Artificial Intelligence and Applications, Vol. 113, pp. 209-216. ISBN: 978-1-58603-466-5. October 2004.
Advertisements