Solving Multilabel MRFs Using Incremental alpha-Expansion on the GPUs

Many vision problems map to the minimization of an energy function over a discrete MRF Fast performance is needed if the energy minimization is one step in a control loop In this paper, we present the incremental α-expansion algorithm for high-performance multilabel MRF optimization on the GPU Our algorithm utilizes the grid structure of the MRFs for good parallelism on the GPU We improve the basic push-relabel implementation of graph cuts using the atomic operations of the GPU and by processing blocks stochastically We also reuse the flow using reparametrization of the graph from cycle to cycle and iteration to iteration for fast performance We show results on various vision problems on standard datasets Our approach takes 950 milliseconds on the GPU for stereo correspondence on Tsukuba image with 16 labels compared to 5.4 seconds on the CPU.

[1]  Donald Geman,et al.  Stochastic relaxation, Gibbs distributions, and the Bayesian restoration of images , 1984 .

[2]  Liang-Gee Chen,et al.  Hardware-Efficient Belief Propagation , 2009, IEEE Transactions on Circuits and Systems for Video Technology.

[3]  Ivan Kovtun,et al.  Partial Optimal Labeling Search for a NP-Hard Subclass of (max, +) Problems , 2003, DAGM-Symposium.

[4]  D. R. Fulkerson,et al.  Flows in Networks. , 1964 .

[5]  Andrew V. Goldberg,et al.  A new approach to the maximum flow problem , 1986, STOC '86.

[6]  Richard J. Anderson,et al.  On the parallel implementation of Goldberg's maximum flow algorithm , 1992, SPAA '92.

[7]  Olga Veksler,et al.  Fast approximate energy minimization via graph cuts , 2001, Proceedings of the Seventh IEEE International Conference on Computer Vision.

[8]  Pushmeet Kohli,et al.  Dynamic Graph Cuts for Efficient Inference in Markov Random Fields , 2007, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[9]  P. J. Narayanan,et al.  CUDA cuts: Fast graph cuts on the GPU , 2008, 2008 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops.

[10]  Olivier Juan,et al.  Active Graph Cuts , 2006, 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06).

[11]  Philip N. Klein,et al.  An O (n log n) algorithm for maximum st-flow in a directed planar graph , 2006, SODA '06.

[12]  Vladimir Kolmogorov,et al.  Convergent Tree-Reweighted Message Passing for Energy Minimization , 2006, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[13]  Axel Pinz,et al.  Computer Vision – ECCV 2006 , 2006, Lecture Notes in Computer Science.

[14]  Richard Szeliski,et al.  A Comparative Study of Energy Minimization Methods for Markov Random Fields , 2006, ECCV.

[15]  Andrew V. Goldberg,et al.  Implementing the Push-Relabel Method for the Maximum Flow Problem on a Connection Machine , 1991, Network Flows And Matching.

[16]  Yuri Boykov,et al.  A Scalable graph-cut algorithm for N-D grids , 2008, 2008 IEEE Conference on Computer Vision and Pattern Recognition.

[17]  Mohamed E. Hussein,et al.  On Implementing Graph Cuts on CUDA , 2007 .

[18]  VekslerOlga,et al.  Fast Approximate Energy Minimization via Graph Cuts , 2001 .

[19]  Pushmeet Kohli,et al.  Reduce, reuse & recycle: Efficiently solving multi-label MRFs , 2008, 2008 IEEE Conference on Computer Vision and Pattern Recognition.

[20]  Olga Veksler,et al.  Fast Approximate Energy Minimization via Graph Cuts , 2001, IEEE Trans. Pattern Anal. Mach. Intell..

[21]  Nikos Komodakis,et al.  Fast, Approximately Optimal Solutions for Single and Dynamic MRFs , 2007, 2007 IEEE Conference on Computer Vision and Pattern Recognition.

[22]  William T. Freeman,et al.  Learning low-level vision , 1999, Proceedings of the Seventh IEEE International Conference on Computer Vision.

[23]  Daniel Cremers,et al.  Efficient planar graph cuts with applications in Computer Vision , 2009, CVPR.