MPLP++: Fast, Parallel Dual Block-Coordinate Ascent for Dense Graphical Models

Dense, discrete Graphical Models with pairwise potentials are a powerful class of models which are employed in state-of-the-art computer vision and bio-imaging applications. This work introduces a new MAP-solver, based on the popular Dual Block-Coordinate Ascent principle. Surprisingly, by making a small change to a low-performing solver, the Max Product Linear Programming (MPLP) algorithm [7], we derive the new solver MPLP++ that significantly outperforms all existing solvers by a large margin, including the state-of-the-art solver Tree-Reweighted Sequential (TRW-S) message-passing algorithm [17]. Additionally, our solver is highly parallel, in contrast to TRW-S, which gives a further boost in performance with the proposed GPU and multi-thread CPU implementations. We verify the superiority of our algorithm on dense problems from publicly available benchmarks as well as a new benchmark for 6D Object Pose estimation. We also provide an ablation study with respect to graph density.

[1]  Bevan K. Youse,et al.  Introduction to real analysis , 1972 .

[2]  Ieee Xplore,et al.  IEEE Transactions on Pattern Analysis and Machine Intelligence Information for Authors , 2022, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[3]  Alok Aggarwal,et al.  Geometric applications of a matrix-searching algorithm , 1987, SCG '86.

[4]  M. I. Schlesinger,et al.  Some solvable subclasses of structural recognition problems , 2000 .

[5]  T. Schiex.  Arc consistency for soft constraints , 2000, Artif. Intell..

[6]  Geir Dahl,et al.  Lagrangian-based methods for finding MAP solutions for MRF models , 2000, IEEE Trans. Image Process..

[7]  Wei Chen,et al.  On Computing the Upper Envelope of Segments in Parallel , 2002, IEEE Trans. Parallel Distributed Syst..

[8]  Alexander Schrijver,et al.  Combinatorial optimization. Polyhedra and efficiency. , 2003 .

[9]  Vladimir Kolmogorov,et al.  Comparison of Energy Minimization Algorithms for Highly Connected Graphs , 2006, ECCV.

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

[11]  Nikos Komodakis,et al.  MRF Optimization via Dual Decomposition: Message-Passing Revisited , 2007, 2007 IEEE 11th International Conference on Computer Vision.

[12]  Tommi S. Jaakkola,et al.  Fixing Max-Product: Convergent Message Passing Algorithms for MAP LP-Relaxations , 2007, NIPS.

[13]  Tomás Werner,et al.  A Linear Programming Approach to Max-Sum Problem: A Review , 2007, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[14]  Yair Weiss,et al.  Minimizing and Learning Energy Functions for Side-Chain Prediction , 2007, RECOMB.

[15]  Richard Szeliski,et al.  A Comparative Study of Energy Minimization Methods for Markov Random Fields with Smoothness-Based Priors , 2008, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[16]  Martin J. Wainwright,et al.  Message-passing for graph-structured linear programs: proximal projections, convergence and rounding schemes , 2008, ICML '08.

[17]  Christoph Schnörr,et al.  A Study of Parts-Based Object Class Detection Using Complete Graphs , 2010, International Journal of Computer Vision.

[18]  Martin C. Cooper,et al.  Soft arc consistency revisited , 2010, Artif. Intell..

[19]  Tamir Hazan,et al.  Norm-Product Belief Propagation: Primal-Dual Message-Passing for Approximate Inference , 2009, IEEE Transactions on Information Theory.

[20]  Martin J. Wainwright,et al.  Message-passing for Graph-structured Linear Programs: Proximal Methods and Rounding Schemes , 2010, J. Mach. Learn. Res..

[21]  Eric P. Xing,et al.  An Augmented Lagrangian Approach to Constrained MAP Inference , 2011, ICML.

[22]  Christoph Schnörr,et al.  A study of Nesterov's scheme for Lagrangian decomposition and MAP labeling , 2011, CVPR 2011.

[23]  Christoph Schnörr,et al.  Evaluation of a First-Order Primal-Dual Algorithm for MRF Energy Minimization , 2011, EMMCVPR.

[24]  M. I. Schlesingera,et al.  Diffusion algorithms and structural recognition optimization problems , 2011 .

[25]  Sebastian Nowozin,et al.  Decision tree fields , 2011, 2011 International Conference on Computer Vision.

[26]  Vladlen Koltun,et al.  Efficient Inference in Fully Connected CRFs with Gaussian Edge Potentials , 2011, NIPS.

[27]  Ofer Meshi,et al.  An Alternating Direction Method for Dual MAP LP Relaxation , 2011, ECML/PKDD.

[28]  Christoph Schnörr,et al.  Efficient MRF Energy Minimization via Adaptive Diminishing Smoothing , 2012, UAI.

[29]  Christoph Schnörr,et al.  A bundle approach to efficient MAP-inference by Lagrangian relaxation , 2012, 2012 IEEE Conference on Computer Vision and Pattern Recognition.

[30]  Berç Rustem,et al.  Solving MRF Minimization by Mirror Descent , 2012, ISVC.

[31]  Tommi S. Jaakkola,et al.  Convergence Rate Analysis of MAP Coordinate Minimization Algorithms , 2012, NIPS.

[32]  Stanislav Zivny,et al.  The Power of Linear Programming for Valued CSPs , 2012, 2012 IEEE 53rd Annual Symposium on Foundations of Computer Science.

[33]  Rob A. Rutenbar,et al.  Hardware implementation of MRF map inference on an FPGA platform , 2012, 22nd International Conference on Field Programmable Logic and Applications (FPL).

[34]  Tomás Werner,et al.  Universality of the Local Marginal Polytope , 2013, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[35]  Sebastian Nowozin,et al.  A Comparative Study of Modern Inference Techniques for Discrete Energy Minimization Problems , 2013, 2013 IEEE Conference on Computer Vision and Pattern Recognition.

[36]  Christoph Schnörr,et al.  Global MAP-Optimality by Shrinking the Combinatorial Search Area with Convex Relaxation , 2013, NIPS.

[37]  Christoph H. Lampert,et al.  1 The Power of LP Relaxation for MAP Inference , 2013 .

[38]  Eugene W. Myers,et al.  Active Graph Matching for Automatic Joint Segmentation and Annotation of C. elegans , 2014, MICCAI.

[39]  Sebastian Nowozin,et al.  A Comparative Study of Modern Inference Techniques for Structured Discrete Energy Minimization Problems , 2014, International Journal of Computer Vision.

[40]  Vladimir Kolmogorov,et al.  The Power of Linear Programming for General-Valued CSPs , 2013, SIAM J. Comput..

[41]  Vladimir Kolmogorov,et al.  A New Look at Reweighted Message Passing , 2015, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[42]  Alan L. Yuille,et al.  Learning Deep Structured Models , 2014, ICML.

[43]  Bogdan Savchynskyy,et al.  Maximum persistency via iterative relaxed inference with graphical models , 2015, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[44]  Rob A. Rutenbar,et al.  Fast hierarchical implementation of sequential tree-reweighted belief propagation for probabilistic inference , 2015, 2015 25th International Conference on Field Programmable Logic and Applications (FPL).

[45]  Christoph Schnörr,et al.  Partial Optimality by Pruning for MAP-Inference with General Graphical Models , 2014, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[46]  Daniel Huber,et al.  Complexity of Discrete Energy Minimization Problems , 2016, ECCV.

[47]  Carsten Rother,et al.  Joint Training of Generic CNN-CRF Models with Stochastic Optimization , 2016, ACCV.

[48]  Thomas Pock,et al.  Solving Dense Image Matching in Real-Time using Discrete-Continuous Optimization , 2016, ArXiv.

[49]  Barry O'Sullivan,et al.  Multi-language evaluation of exact solvers in graphical model discrete optimization , 2016, Constraints.

[50]  Eric Brachmann,et al.  Global Hypothesis Generation for 6D Object Pose Estimation , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[51]  Nikos Komodakis,et al.  Newton-Type Methods for Inference in Higher-Order Markov Random Fields , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[52]  Tomás Werner,et al.  LP Relaxation of the Potts Labeling Problem Is as Hard as Any Linear Program , 2017, IEEE Transactions on Pattern Analysis and Machine Intelligence.