The Community for Technology Leaders
Green Image
Issue No. 03 - March (2009 vol. 31)
ISSN: 0162-8828
pp: 492-504
Qingxiong Yang , University of Illinois at Urbana Champaign, Urbana
Liang Wang , University of Kentucky, Lexington
Ruigang Yang , University of Kentucky, Lexington
Henrik Stewénius , Google Switzerland, Zurich
David Nistér , Microsoft Corp., Redmond
In this paper, we formulate a stereo matching algorithm with careful handling of disparity, discontinuity and occlusion. The algorithm works with a global matching stereo model based on an energy-minimization framework. The global energy contains two terms, the data term and the smoothness term. The data term is first approximated by a color-weighted correlation, then refined in occluded and low-texture areas in a {\it{repeated}} application of a hierarchical loopy belief propagation algorithm. The experimental results are evaluated on the Middlebury data sets, showing that our algorithm is the {\it{top}} performer among all the algorithms listed there.
3D/stereo scene analysis, Segmentation, Markov random fields

R. Yang, H. Stewénius, Q. Yang, D. Nistér and L. Wang, "Stereo Matching with Color-Weighted Correlation, Hierarchical Belief Propagation, and Occlusion Handling," in IEEE Transactions on Pattern Analysis & Machine Intelligence, vol. 31, no. , pp. 492-504, 2008.
82 ms
(Ver 3.3 (11022016))