PMBP: PatchMatch Belief Propagation for Correspondence Field Estimation

被引:243
作者
Besse, Frederic [1 ]
Rother, Carsten [1 ]
Fitzgibbon, Andrew [1 ]
Kautz, Jan [1 ]
机构
[1] UCL, London, England
关键词
Correspondence fields; Belief propagation; PatchMatch;
D O I
10.1007/s11263-013-0653-9
中图分类号
TP18 [人工智能理论];
学科分类号
140502 [人工智能];
摘要
PatchMatch (PM) is a simple, yet very powerful and successful method for optimizing continuous labelling problems. The algorithm has two main ingredients: the update of the solution space by sampling and the use of the spatial neighbourhood to propagate samples. We show how these ingredients are related to steps in a specific form of belief propagation (BP) in the continuous space, called max-product particle BP (MP-PBP). However, MP-PBP has thus far been too slow to allow complex state spaces. In the case where all nodes share a common state space and the smoothness prior favours equal values, we show that unifying the two approaches yields a new algorithm, PMBP, which is more accurate than PM and orders of magnitude faster than MP-PBP. To illustrate the benefits of our PMBP method we have built a new stereo matching algorithm with unary terms which are borrowed from the recent PM Stereo work and novel realistic pairwise terms that provide smoothness. We have experimentally verified that our method is an improvement over state-of-the-art techniques at sub-pixel accuracy level.
引用
收藏
页码:2 / 13
页数:12
相关论文
共 20 条
[1]
Achan K., 2008, ADV NEURAL INFORM PR, P737
[2]
Barnes C., 2010, Proc. ECCV
[3]
PatchMatch: A Randomized Correspondence Algorithm for Structural Image Editing [J].
Barnes, Connelly ;
Shechtman, Eli ;
Finkelstein, Adam ;
Goldman, Dan B. .
ACM TRANSACTIONS ON GRAPHICS, 2009, 28 (03)
[4]
Bleyer M., 2011, P BMVC
[5]
RANDOMIZED MOTION ESTIMATION [J].
Boltz, Sylvain ;
Nielsen, Frank .
2010 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING, 2010, :781-784
[6]
Non-Rigid Dense Correspondence with Applications for Image Enhancement [J].
HaCohen, Yoav ;
Shechtman, Eli ;
Goldman, Dan B. ;
Lischinski, Dani .
ACM TRANSACTIONS ON GRAPHICS, 2011, 30 (04)
[7]
Ihler A. T., 2009, AISTATS, P256
[8]
Kaiming He, 2011, 2011 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), P2049, DOI 10.1109/CVPR.2011.5995495
[9]
Convergent tree-reweighted message passing for energy minimization [J].
Kolmogorov, Vladimir .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2006, 28 (10) :1568-1583
[10]
Korman S, 2011, IEEE I CONF COMP VIS, P1607, DOI 10.1109/ICCV.2011.6126421