Vision and Robotics Seminar
will speak on
Multi-body Segmentation: Revisiting Motion Consistency
Dynamic analysis of video sequences often relies on the segmentation of the sequence into regions of consistent motions. Approaching this problem requires a definition of which motions are regarded as consistent. Common approaches to motion segmentation usually group together points or image regions that have the same motion between successive frames (where the same motion can be 2D, 3D, or non-rigid). We define a new type of motion consistency, which is based on temporal consistency of behaviors across multiple frames in the video sequence. Our definition of consistent ``temporal behavior" is expressed in terms of multi-frame linear subspace constraints. This definition applies to 2D, 3D, and some non-rigid motions without requiring prior model selection. We further present a multi-frame multi-body segmentation algorithm which applies the new motion consistency constraint directly to image brightness measurements, without requiring prior correspondence estimation nor feature tracking.
This is joint work with Lihi Zelnik-Manor and Michal Irani.
The lecture will take place in the
Lecture Hall, Room 1, Ziskind Building
on Thursday, May 2, 2002