Research


Data Synthesis

We tackle the problem of full-body 3D pose estimation in the wild. To address the lack of training data, we augment a dataset of real images with new synthetic images and showed that Convolutional Neural Networks (CNN) can be trained on artificially-looking images and generalize well to real images.

 
Click here for more details…
 


Egovision4health

Egovision4Health is a joint research project between the University of Zaragoza, Spain and the University of California, in Irvine, USA. The objective of this three-year project is to investigate new egocentric computer vision techniques to automatically provide health professionals with an assessment of their patients’ ability to manipulate objects and perform daily activities.

 
Click here for more details…
 


Human Pose Detection

We address human detection and pose estimation from monocular images by formulating  it as a classification problem. Our main contribution is a multi-class pose detector that uses the best components of state-of-the-art classifiers including hierarchical trees, cascades of rejectors as well as randomized forests.

 
Click here for more details…
 


View-invariant Tracking in Video-surveillance

We present a methodology for viewpoint invariant monocular 3D human pose tracking in man-made environments in which we exploit some properties of projective geometry and assume that observed people move on a known ground plane.

 
Click here for more details…
 

 

 


Spatio-Temporal Framework

We propose a model-based approach associating the body pose and the 2D silhouette to jointly segment and recover the pose of the subject observed in the scene. Local spatio-temporal models corresponding to several views and steps of the same action are trained, concatenated and sorted in a global framework.

 
Click here for more details…


RBPF for 2D Pose Tracking

We propose to exploit spatio-temporal constraints in a Rao-Blackwellized Particle Filter framework for robust 2D pose tracking.

 
Click here for more details…
 

 

 

 


OCR

This R&D project started as an Optical Characters Recognition (OCR) system for automatic Iberian ham traceability. It was followed by a R&D project with the company DeInta SL to develop a car license plate recognition system.

 
Click here for more details…