M. Nicolescu and G. Medioni (USA)
Stereo vision, motion segmentation
A main difficulty for estimating camera and scene geometry from a set of point correspondences is caused by the presence of false matches and independently moving objects. Given two images, after obtaining the matching points they are usually filtered by an outlier rejection step before being used to solve for epipolar geometry and 3-D structure estimation. In the presence of moving objects, image registration becomes a more challenging problem, as the matching and registration phases become interdependent. We propose a novel approach that decouples the above operations, allowing for explicit and separate handling of matching, outlier rejection, grouping, and recovery of camera and scene structure. The key aspect is that we first determine an accurate representation in terms of dense image velocities (equivalent to point correspondences), segmented motion regions and boundaries, by using only the smoothness of image motion. Only then we proceed with the extraction of camera and scene 3-D geometry, separately for each rigidly moving object.
Important Links:
Go Back
