VISTA - Spatio-Temporal Vision

IRISA / INRIA Rennes - Bretagne Atlantique
Campus universitaire de Beaulieu
35042 RENNES Cedex, France


July 1, 2009: We are very sad to announce that our friend and colleague Jean-Pierre Lecadre passed away

April 10, 2009: Minisymposium on "Patch-based Image Representation, Manifolds and Sparsity" (slides and videos)

January 19, 2009: we are currently recruiting 2 postdocs. See details.

 
Vista research work is concerned with various types of spatio-temporal images (mainly video images, but also video-microscopy and X-ray images). We are investigating methods to analyze dynamic scenes, and, more generally, dynamic phenomena, within image sequences. We address the full range of problems raised by the analysis of such dynamic contents with a focus on image motion analysis issues: space-time denoising, motion detection, motion estimation, motion-based segmentation, tracking, motion recognition and interpretation with learning. We usually rely on statistical approaches, resorting to: Markov models, Bayesian inference, robust estimation, particle filtering, learning. Application-wise, we focus our attention on two main domains: content-aware video applications and
biological imaging. For that, a number of collaborations, academic and industrial, national and international, are set up.

Key-words : image sequence processing, motion analysis, tracking and trajectography, detection, measurement, denoising, learning and recognition of spatio-temporal events, video processing, biological imagery.








 

  Project leader: Patrick Pérez


  Webmaster : vista_webmaster@irisa.fr
  Last modified: 18-09-2009