Solving Stereo Correspondence through Minimizing Energy Function with Higher-Order Cliques

摘要：

Stereo correspondence is one of the most active research areas in computer vision.Energy minimization is widely used for early vision problems,such as image restoration, segmentation and stereo correspondence.Pairwise clique is the most commonly used smoothness term of energy function, but it is unable to capture rich statistics of natural scene. Energy function considering higher-order clique potentials can characterizes richer statistics of natural scene than pairwise clique,but it is di .cult to model higher-order clique potentials and the computation for minimization is much heavier. We introduce an reduced P n Potts model which can char-acterize higher-order clique potentials and was .rst used for image segmentation.Speci .cally,we present two new models which map the Pn Potts model to α -expansion move and α -βswap move.Furthermore,we propose a new graph construction method for them which has fewer extra nodes than before.Those models can be easily applied to other vision problems.The experiment shows that the results considering P n Potts model are more accurate than those without.

关键词： Stereo correspondence graph cuts pairwise higher-order cliques disparity maps

作者: Guowei Wan Aiping Wang Sikun Li Liang Zeng

作者单位: School of Computer Science National University of Defense technology Changsha,Hunan Province,China

会议类型: 国际会议

会议名称: 2008 IEEE International Conference on Onformation and Automation(IEEE 信息与自动化国际会议)

会议地点: 张家界

会议语种:英文

页码: 407-412

在线出版日期: 2008-06-20（万方平台首次上网日期，不代表论文的发表时间）