Dense Motion and Disparity Estimation Via Loopy Belief Propagation

We describe a method for computing a dense estimate of motion and disparity, given a stereo video sequence containing moving non-rigid objects. In contrast to previous approaches, motion and disparity are estimated simultaneously from a single coherent probabilistic model that correctly accounts for all occlusions, depth discontinuities, and motion discontinuities. The results demonstrate that simultaneous estimation of motion and disparity is superior to estimating either in isolation, and show the promise of the technique for accurate, probabilistically justified, scene analysis. 1 Motivation and previous work The “temporal stereo + motion” problem of estimating the disparity and motion fields in a video sequence of moving objects captured by a calibrated pair of stereo cameras has been studied for at least two decades [1]. It is worthwhile to distinguish between the standard temporal stereo + motion problem, and the more restricted problem of estimating disparity and motion fr...

• Type: Conference
• Year: 2006
• Where: ACCV
• Authors: Michael Isard, John MacCormick
• Keywords: ACCV 2006, Computer Vision, Stereo, Stereo Sequence, Temporal Stereo