Human Pose Recovery and Behavior Analysis

Human Action/Gesture recognition is a challenging area of research that deals with the problem of recognizing people in images, detecting and describing body parts, including hand pose and facial expressions, inferring their spatial configuration, and performing action/gesture recognition from still images or image sequences, also including multi-modal data. Because of huge space of human configurations, body pose recovery is a difficult problem that involves dealing with several distortions: illumination changes, partial occlusions, changes in the point of view, rigid and elastic deformations, or high inter and intra-class variability, just to mention a few. Even with the high difficulty of the problem, modern Computer Vision techniques and new tendencies deserve further attention, and promising results are expected in the next years. Moreover, several subareas have been recently defined, such as Affective Computing, Personality Computing, Social Signal Processing, Human Behavior Analysis, or Social Robotics. The effort involved in this area of research will be compensated by its potential applications: social robotics, TV production, home entertainment (multimedia content analysis), education purposes, sociology research, surveillance and security, improved quality live by means of monitoring or automatic artificial assistance, etc.


HuPBA Computer Vision Center Webpage



Group members



Marc Oliu received the Technical Bachelor degree in Computer Science from Universitat de Girona, Girona, in 2010. He is finishing his Master degree in Artificial Intelligence at Universitat Politècnica de Catalunya (UPC), Universitat de Barcelona (UB), and Universitat Rovira  i Virgili (URV). He is interested in Computer Vision and Machine Learning Fields.

Master Thesis: Head pose recovery and shape estimation in still images, UB-URV-UPC, 2014
Presentation slides
Email: moliusimon (at) gmail (dot) com



Ciprian Adrian Corneanu  got his MSc in Telecommunication Engineering from Télécom SudParis in 2011. He spent the next three years in Germany with C.R.S iiMotion, a Thomson spin-off, developing industrial image processing applications for the consumer market. Currently he is a Ph.D student at the Universitat de Barcelona and a fellow of the Computer Vision Center from Universitat Autonoma de Barcelona. His main research interests include face and behavior analysis, affective computing, social signal processing, human computer interaction.

Master Thesis: Facial Expression Analysis of Neurologically Impaired Children, CVC, 2015. Slides
Email: cipriancorneanu (at) gmail (dot) com Personal webpage



Albert Clapés received his Bachelor degree in Computer Science at Universitat de Barcelona (UB) in 2012. He obtained the Master degree in Artificial Intelligence in 2014 at UPC (Universitat Politècnica de Catalunya), UB and URV. He did the PhD on action recognition.His reserch interests include computer vision applied to human behavior analysis and human-computer interaction.

Master Thesis: A flexible descriptive framework for the visual analysis of humans: a modality-independent approach, 2014.
PhD Thesis: Learning to recognize human actions: from hand-crafted to deep-learning based visual representations, 2019. Slides
Email: aclapes (at) cvc (dot) uab (dot) cat



Daniel Sánchez received his Bachelor degree in Computer Science at Universitat de Barcelona (UB) in 2012. He is currently studying his Master degree in Artificial Intelligence at UPC (Universitat Politècnica de Catalunya), UB and URV. He is mainly interested in computer vision applied to human pose recovery and behavior analysis.

Master Thesis: ECOC and Graph Cuts Segmentation of Human Limbs, UB-URV-UPC, 2014. Slides
Email: gammarl (at) gmail (dot) com



Cristina Palmero received her Bachelor degree in Audiovisual Telecommunication Systems Engineering at Universitat Politècnica de Catalunya (UPC) in 2011. She is currently studying her Master degree in Artificial Intelligence at Universitat de Barcelona (UB), Universitat Politècnica de Catalunya (UPC) and Universitat Rovira  i Virgili (URV). She is mainly interested in signal and image processing and computer vision techniques applied  to human behavior analysis, scene understanding and robotics.

Master Thesis: Trimodal Human Body Segmentation, UB-URV-UPC, 2014.
Master Thesis slides
Email: c (dot) palmero (dot) cantarino (at) gmail (dot) com



Meysam Madadi received his Bachelor degree in Software Engineering at BuAli Sina university of Hamedan and M.S. degree in Computer Vision and Artificial Intelligence at Universitat Autònoma de Barcelona (UAB) in 2007 and 2013, respectively. He did research on human pose recovery and human behavior analysis for his master thesis Extraction of body soft-biometry from 3D videos using Kinect. He is interested in generating and developing new algorithms in these topics applying the knowledge in computer vision and retrieval systems besides machine learning, algorithms design in artificial intelligence, statistics, and linear algebra.

PhD Thesis: Human Segmentation, Pose Estimation and Applications, Universitat Autonoma de Barcelona, 2017. Slides
Email: meysam (dot) madadi (at) gmail (dot) com



Julio C. S. Jacques Junior received his MS degree in Applied Computing in 2006, from UNISINOS and his Ph.D. degree in Computer Science in 2012, from PUCRS, Brazil. Currently, he is a postdoctoral researcher at the Computer Science, Multimedia and Telecommunications department at Universitat Oberta de Catalunya (UOC), within the Scene Understanding and Artificial Intelligence (SUnAI) group. He is also a researcher collaborator within the Computer Vision Center (CVC) and Human Pose Recovery and Behavior Analysis group (HuPBA) at Universitat Autonoma de Barcelona (UAB) and University of Barcelona (UB), as well as within ChaLearn Looking at People. His research interests include, between others, computer vision-based applications with special attention in visual human behavior analysis.

Email: juliojj (at) gmail (dot) com      Personal webpage



Hugo Bertiche Argila received his Bachelor degree in Aeronautical Engineering at Universitat Politècnica de Catalunya (UPC) in 2015. Afterwards, he obtained the Master degree in Artificial Intelligence in 2017 at UPC, Universitat de Barcelona (UB) and Universitat Rovira i Virgili (URV). He is currently a PhD student at UB. His main research line is deep learning for human pose and shape recovery.

Master Thesis: Comparison of motion-based approaches for multi-modal action and gesture recognition from RGB-D, UPC-URV-Universitat de Barcelona, 2017. Slides.
Email: hugo_bertiche (at) hotmail (dot) com



Javier Selva Castelló received his Bachelor degree in Computer Science at Universitat Politècnica de València (UPV) in 2015. Two years later he got his MSc in Artificial Intelligence from Universitat de Barcelona (UB), Universitat Politècnica de Catalunya (UPC) and Universitat Rovila i Virgili (URV). Currently he is a Ph.D student at the Universitat de Barcelona. His main research interests include Machine Learning, Natural Language Processing and Computer Vision for human understanding and behaviour analysis.

Master Thesis: A Comprehensive Survey on Deep Future Frame Video Prediction, UB-UPC-URV, 2018. Slides
Email: javier (dot) selvac (at) gmail (dot) com


Past students and members

Mehdi Mirza-Mohammadi Master Thesis: Contextual Bag-of-Visual-Words and ECOC-Rank for Retrieval and Multi-class Object Recognition. Universitat de Barcelona-URV-UPC, 2009

2009- Alberto Escudero – Research assistant

Oscar Lopes Master Thesis: Real-time Hand Pose Recognition using Depth Sensors combined with Spherical Shape Model Descriptor. Slides. Computer Vision Center, 2012

Miguel Pousa Master Thesis: Spherical Blurred Shape Model for Hand Pose Recognition. Computer Vision Center, 2012

2009-2013 – Mario Gorga – Research assistant

2013 – Albert Marieges – Research assistant 

2014 – Vicent Roig – Research assistant

2014 – Guillermo Blasco – Research assistant

2014 – David Corto – Research assistant

2014 – Andres Cenci – Research assistant

Eloi Puertas PhD Thesis: Generalized Stacked Sequential Learning. Slides. Universitat de Barcelona, 2014

Carlos Puig Master Thesis: Automatic Performance Analysis in Trampoline from RGB-Depth Data. Presentation. Poster. Video Sample. Computer Vision Center, 2014

Oscar Amoros Master Thesis: Parallel error-correcting output codes classification in volume visualization: parallelism for AI and AI for parallelism. Slides. Universitat de Barcelona-URV-UPC, 2014

2015 – Lukasz Romaszko – Research assistant

Mikkel Thøgersen Master Thesis: Segmentation of rgb-d indoor scenes by stacking random forests and conditional random fields. Slides., Aalborg University, 2015

Pablo Pardo Master Thesis: Automatic Real and Apparent Age Estimation in Still Images. Slides., UPC-URV-Universitat de Barcelona, 2015

2015 – Marc Pomar – Research assistant

2015 – 2016 Sergi Delgado – Research assistant

Xavier Perez-Sala Master Thesis: Vision-based Navigation and Reinforcement Learning Path Finding for Social Robots. Slides. UPC-URV-Universitat de Barcelona, 2010. PhD Thesis: Extending Procrustes Analysis: Building Multi-view 2-D Models from 3-D Human Shape Samples. Universitat Politecnica de Catalunya, 2015. Email: xperez (at) andrew (dot) cmu (dot) edu.

Antonio Hernández-Vela Master Thesis: Pose and Face Recovery via Spatio-temporal GrabCut Human Segmentation, Computer Vision Center, 2010. PhD Thesis: From pixels to gestures: learning visual representations for human analysis in color and depth data sequences. Universitat de Barcelona, 2015. Slides. Personal webpage. Email: chesterr66 (at) gmail (dot) com

Miguel Ángel Bautista Master Thesis: Sublinear Evolutionary Design of Error-Correcting Output Codes Best Catalonia Prize by the Catalan Association on Artificial Intelligence. Master presentation slides, UPC-URV-Universitat de Barcelona, 2011. PhD Thesis: Learning Error-Correcting Representations for Multi-Class Problems, Universitat de Barcelona, 2015. Slides. Personal webpage Best PhD Award

Gerard Canal Camprodon Master Thesis: Human multi-robot interaction based on gesture recognition. UPC-URV-Universitat de Barcelona Slides Best Catalonia Prize by the Catalan Association on Artificial Intelligence Email: gcanalcamprodon (at) gmail (dot) com, 2015

Pouya Ahmadmonfared Master Thesis: RGB-D Graph Based Human Body Segmentation for Mattress Recommendation, Computer Vision Center, 2016

Mohammad Ali Bagheri PhD Thesis: Ensemble Learning for Visual Recognition. Dalhousie University, 2016. Slides

Marco Bella Master Thesis: Hybrid CNN+LSTM for Face Recognition in Videos. UPC-URV-Universitat de Barcelona, 2016. Slides

Miguel Reyes Master Thesis: Human Pose Recovery and Behavior Analysis from RGB and Depth Maps, 2011. PhD Thesis: Human Pose Analysis and Gesture Recognition from Depth Maps: Methods and Applications, 2016. Slides. Personal webpage Email: mreyese (at) gmail (dot) com

2016-2017 – Maryam Asadi – PhD research stay

2016-2017 – Oriol Suils – Research assistant

Xavier Serra Master Thesis: Face recognition using Deep Learning, UPC-URV-Universitat de Barcelona, 2017. Slides

Andrei Polzounov Master Thesis: WordFences: Text Localization and Recognition, UPC-URV-Universitat de Barcelona, 2017. Slides

Víctor Ponce-López Master Thesis: Multi-modal Human Gesture Recognition combining Dynamic Programming and Probabilistic Methods. Presentation slides, 2012. PhD Thesis: Evolutionary Bags of Space-time Features for Human Analysis. 2016. Slides. Personal webpage, Email: vponcelop (at) gmail (dot) com

Frederic Sampedro PhD Thesis: Automatic image quantification strategies in clinical nuclear medicine and neuroradiology. Universitat Autonoma de Barcelona. 2017. Slides. Personal webpage, Email: fredsampedro (at) gmail (dot) com Best PhD Award

Egils Avots Master Thesis: Garment retexturing using Kinect V2.0, University of Tartu, Robotics and Computer Engineering, 2017. Slides.

Valentin Ramirez Master Thesis: First Approach on Automated Psychomorphological Analysis Implementation via Computer Vision, Universitat de Barcelona, 2017.

Deividas Skiparis Master Thesis: Semantic face segmentation from video streams in the wild, UPC-URV-Universitat de Barcelona, 2017. Slides.

2017 – Julien Bonnier – Research stay

2017 – Mohamed Llyes – Research stay

2017 – Ruben Ballester – Research assistant

Gonzalo Benito Master Thesis: Comparing classical and deep approaches for face recognition in a smartgym application, CVC-UAB-UPC-UPF, 2017. Slides.

Vicent Roig Master Thesis: Multimodal 2DCNN action recognition from RGB-D Data with Video Summarization, UPC-URV-Universitat de Barcelona, 2017. Slides.

Martí Soler Master Thesis: Recovering missing data from human body images, 2018. Slides.

Juanjo Rubio Guillamon Master Thesis: Deep Learning for Infrastructure Damage Categorization, 2018.

Daniel Natanael Garca Zapata Master Thesis: Deep Learning for Expression Recognition in Image Sequences, 2018. Slides.

Juan Luis Rosa Ramos Master Thesis: Deep Learning for Universal Emotion Recognition in Still images, 2018. Slides.

Juan Borrego Master Thesis: Fully Convolutional Architectures for Multi-Part Body Segmentation, 2018. Slides.

Xenia Salinas Master Thesis: Spatio-Temporal Emotion Recognition, 2018. 2018 Research assistant

2018 – Swathikiran Sudhakaran – Research stay

2018 – Zhengying Liu – Research stay

2018 – Pau Buch-Cardona – Research assistant

Ricardo Darío Pérez Master Thesis: Multiple attribute analysis for deep learning based apparent personality analysis, 2018

2019 – Noor Ul Huda – Research stay

Pau Bramon Master Thesis: Deep 3D Pose Regression of Real Objects, 2019. Slides.

Santiago Mazagatos Master Thesis: Multi-Scale Super Resolution With Blind De-noising Using Residual Learning For Digital Art, 2019. Slides.

Pablo Arcos Master Thesis: Virtual Try-on with Generative Adversarial Networks, 2019. Slides.