Organisers: A.I. Comport, A. Bartoli (CNRS-LASMEA, France)

Contact: This e-mail address is being protected from spambots. You need JavaScript enabled to view it , This e-mail address is being protected from spambots. You need JavaScript enabled to view it

Lecturers: Adrien Bartoli, Selim Benhimane, Andrew Comport, Kurt Konolige, Vincent Lepetit, Philippe Martinet, Walterio Mayol-Cuevas

Tutorial Date: 22nd of September 2008

NEW

Thanks to all the speakers and participants for this successful tutorial. Please find here the presentations of each speaker:

Summary

In this  full day tutorial, we aim to give a survey of recent developments in computer vision for robotics applications.

In the first half of the tutorial a unifying framework will be presented which situates robotics control concepts within the context of computer vision tracking algorithms. Then, we will see how computer vision and machine learning techniques can be combined in order provide robust and efficient feature detection style tracking. Next a wireframe model based localisation technique will be presented and related to a multi-view visual odometry technique without any any a-priori structure.

In the second half of the tutorial, visual SLAM techniques will be aborded within the context of spatially aware robots. Then several challenging outdoor SLAM problems will be investigated touching aspects such as scene registration, navigation, place recognition and visual odometry. Finally, after having considered the camera environment as rigid, some advanced computer vision results will be presented based on non-rigid deformable surfaces giving interesting perspectives in robotics.