CMTech is a research group in Computer Vision & Human-Computer Interaction focused on human behavior understanding and interaction with the world.

Human behavior understanding implies the analysis of different cues related with human non-verbal communication. One of the goals of CMTech is to allow computers to be able to understand these cues using Computer Vision and machine learning. CMTech deals with 3 main challenging tasks involving human body and hand pose inference and emotional analysis using facial expressions. These technologies provide solutions in different applications such as postural correction, understanding of sign languages or affective based human-computer interaction.

The HCI branch of CMTech explores the possibilities of Full-Body Interaction based on current Embodied Cognition theories in areas such as Learning, Play and Special Needs. This research is undertaken at the Full-Body Interaction Lab (FubIntLab) from the standpoint of Interaction Design, focusing our attention mainly on interaction for children and using non-invasive technologies such as those mentioned above. To check the research projects within the FuBIntLab check here: FuBIntLab Research Projects.

Research projects

Body activity analysis
The body is involved in most of the human activities. In order to understand automacally the human behaviour, we need to create intelligent systems which understand the motion of the body parts and their relations.

Consumedia
Proyecto INNPACTO 2011 CONSUMEDIA en el cual participa la UNIVERSITAT POMPEU FABRA cofinanciado por el MINECO y FEDER El Ministerio de Economía y Competitividad español (MINECO), dentro del Plan Nacional de Investigación Científica, Desarrollo e...

CRAFTI
"crafti" is the Catalan acronym for "Method for the Control and Regulation of Physical Activity of Children through a Novel Platform for Full-body Interactive Experiences called the Interactive Slide". The goal of this research is to provide...

DLIG: Direct Local Indirect Global alignment for Video Mosaicing
In this article we present a framework for real-time mosaicing from video sequences recorded from an uncalibrated PTZ camera based on multiframe registration. To this end, a new frame alignment algorithm, the Direct Local Indirect Global (DLIG),...

Emotional analysis on facial expressions
Facial expressions are one of the most important channels of communication of the human being. Automatic Facial Expression Analysis it’s a challenging task which tries to allow the computers to analyse and understand these expressions. This...

From Emotions to Action Units with Hidden and Semi-Hidden-Task Learning
Limited annotated training data is a challenging problem in Action Unit recognition. In this paper, we investigate how the use of large databases labelled according to the 6 universal facial expressions can increase the generalization ability of...

Full Body Interaction Learning Environments
Research in cognitive science - showing the embodied nature of cognition - suggests that, from a theoretic perspective, embodied interaction has the potential to support learning. Within this field promising possibilities can be found in learning...

Human 3D modeling
The capture of 3D information is getting a high importance during the last years. We propose a fast system in order to obtain the 3D model of a person using only one single 3D camera and 2 mirrors. This disposition allows the obtention of the...

Human body pose inference from monocular images
Not only detecting the presence of persons on an image, but also estimating their body pose configuration is a challenging task, even more when only monocular images are available, involving the lack of contextual cues such as motion or time...

PROBADOR VIRTUAL 3D 360°: VISUALIZACIÓN 3D DE LA ROPA EN 360 GRADOS
El Ministerio Economía y Competitividad (MINECO), dentro del Plan Nacional de Investigación Científica, Desarrollo e Innovación Tecnológica 2008-2011, concedió ayuda al proyecto liderado por ARTIFICIAL INTELLIGENCE TECHNOLOGIES, S.L. con número de......

RMC-MIL for weakly-supervised facial behavior analysis
Regularized Multi-Concept MIL for weakly-supervised facial behavior categorization Adrià Ruiz, Joost Van de Weijer and Xavier Binefa

Speech analysis
Spoken term detection search into acoustic documents for specific acoustic patterns. In our query by example approach We focus in a totally unsupervised search of an uttered acoustic pattern inside a bigger acoustic document. We present a second...

TFG: A Kinect extension for full 3D human body completion

TFG: Emotion Clustering: Anàlisi no supervisat d’expressions facials
L’anàlisis automàtic d’expressions facials a partir d’imatges està sent un camp de recerca molt actiu en els últims anys. Si els ordinadors fossin capaços d’entendre les expressions facials humanes permetria innumerables possibles aplicacions:...

TFG: Face transfer: animació d’avatars facials.
Mitjançant l’aprenentatge de models facials (avatars) i la detecció de punts facials d’un altre usuari, s’implementarà una tècnica coneguda com a “face transfer” mitjançant la qual es poden animar els models apresos.

Virtual Billiard
The virtual billiard table was a gaming project we developed by combining classical physical games with Computer Vision technology in order to provide a whole new experience to users.