Séminaire (Organisé par l’Equipe de recherche DI)

Philippe XU


Local Information Fusion for Scene Understanding

Mecredi 16 mai 2012 à 14h en salle RD134

Résumé :

Many perception algorithms have been developed over the years in the field of robotics.

Various sensors (mono/stereo-cameras, laser, radar, etc.) and classes of objects (pedestrian, car, road, moving object, etc.) have been considered. Each individual sensor/algorithm has its own representation space (2D grid map, object level space, pixel space, 3D space, etc.) and its own limited perception capability. Laser sensor can, for example, only see the presence or not of obstacles without recognizing different classes of obstacles. A classical image-based pedestrian detector will output a bounding box in the image space and no other class will be considered.

The fusion of such different kind of information is a very challenging task but also crucial to make the most of all the sensors and methods. We propose to reason over cells coming from an over-segmented image in which information will be fused. Each sensor having its own partial view of the world, the belief function theory will be particularly adapted to this case. In this talk, a first set of methods will be considered to illustrate the fusion aspect, laser-based free space detection, stereo-based road/obstacle detection, mono-camera based temporal propagation.


Mardi 20 juin 2017

Séminaire à 14h en GI042 (bâtiment Blaise Pascal), présenté par Patrice Perny, LIP6.
« Décision interactive sur domaine combinatoire par élicitation incrémentale de préférences ».

Jeudi 11 mai 2017

Séminaire à 14h en GI042 (bâtiment Blaise Pascal), présenté par Nicolas Maudet, LIP6 (Equipe SMA).
« Current issues in argumentation ».

Mardi 4 avril 2017

Séminaire à 14 h dans l’amphi du Centre d’Innovation de l’UTC, présenté par Xavier LAGORCE, PhD, Head of Computer Vision, Chronocam.
« Chronocam : Event-based cameras for machine vision »

Jeudi 27 octobre 2016

Séminaire à 14 h en GI016, présenté par Fabien Pfaender, UTSEUS.
« State of Cities - A Massive, Systematic, Data Powered, Comparative Analysis Of Cities ».

Pages 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | ...


FR SHIC 3272

Collegium UTC/CNRS