Abstract
Despite the availability of many Markov Random Field (MRF) optimization algorithms, their widespread usage is currently limited due to imperfect MRF modelling arising from hand-crafted model parameters and the selection of inferior inference algorithm. In addition to differentiability, the two main aspects that enable learning these model parameters are the forward and backward propagation time of the MRF optimization algorithm and its inference capabilities. In this work, we introduce two fast and differentiable message passing algorithms, namely, Iterative Semi-Global Matching Revised (ISGMR) and Parallel Tree-Reweighted Message Passing (TRWP) which are greatly sped up on a GPU by exploiting massive parallelism. Specifically, ISGMR is an iterative and revised version of the standard SGM for general pairwise MRFs with improved optimization effectiveness, and TRWP is a highly parallel version of Sequential TRW (TRWS) for faster optimization. Our experiments on the standard stereo and denoising benchmarks demonstrated that ISGMR and TRWP achieve much lower energies than SGM and Mean-Field (MF), and TRWP is two orders of magnitude faster than TRWS without losing effectiveness in optimization. We further demonstrated the effectiveness of our algorithms on end-to-end learning for semantic segmentation. Notably, our CUDA implementations are at least 7 and 700 times faster than PyTorch GPU implementations for forward and backward propagation respectively, enabling efficient end-to-end learning with message passing.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Hirschmuller, H.: Stereo processing by semiglobal matching and mutual information. TPAMI 30, 328–341 (2008)
Boykov, Y., Jolly, M.: Interactive graph cuts for optimal boundary and region segmentation of objects in N-D images. In: ICCV (2011)
Hassner, M., Sklansky, J.: The use of Markov random fields as models of texture. Comput. Graph. Image Process. 12(4), 357–370 (1980)
Szeliski, R., et al.: A comparative study of energy minimization methods for Markov random fields with smoothness-based priors. TPAMI 30, 1068–1080 (2008)
Zheng, S., et al.: Conditional random fields as recurrent neural networks. In: CVPR (2015)
Krähenbühl, P., Koltunz, V.: Efficient inference in fully connected CRFs with Gaussian edge potentials. In: NeurIPS (2011)
Seki, A., Pollefeys, M.: SGM-Nets: semi-global matching with neural networks. In: CVPR (2017)
Drory, A., Haubold, C., Avidan, S., Hamprecht, F.A.: Semi-global matching: a principled derivation in terms of message passing. In: Proceedings of German Conference on Pattern Recognition (GCPR) (2014)
Kolmogorov, V.: Convergent tree-reweighted message passing for energy minimization. TPAMI 28, 1568–1583 (2006)
Ajanthan, T., Hartley, R., Salzmann, M.: Memory efficient max-flow for multi-label submodular MRFs. In: CVPR (2016)
Ajanthan, T., Hartley, R., Salzmann, M., Li, H.: Iteratively reweighted graph cut for multi-label MRFs with non-convex priors. In: CVPR (2015)
Boykov, Y., Veksler, O., Zabih, R.: Fast approximate energy minimization via graph cuts. TPAMI 23, 1222–1239 (2001)
Carr, P., Hartley, R.: Solving multilabel graph cut problems with multilabel swap. In: DICTA (2009)
Hartley, R., Ajanthan, T.: Generalized range moves. arXiv:1811.09171 (2018)
Veksler, O.: Multi-label moves for MRFs with truncated convex priors. IJCV 98, 1–14 (2012). https://doi.org/10.1007/s11263-011-0491-6
Jordan, M.: Learning in Graphical Models. MIT Press, Cambridge (1998)
Kwon, D., Lee, K.J., Yun, I.D., Lee, S.U.: Solving MRFs with higher-order smoothness priors using hierarchical gradient nodes. In: Kimmel, R., Klette, R., Sugimoto, A. (eds.) ACCV 2010. LNCS, vol. 6492, pp. 121–134. Springer, Heidelberg (2011). https://doi.org/10.1007/978-3-642-19315-6_10
Murphy, K., Weiss, Y., Jordan, M.: Loopy belief propagation for approximate inference: an empirical study. In: UAI (1999)
Pearl, J.: Probabilistic Reasoning in Intelligent Systems. Morgan Kaufmann, Burlington (1988)
Wainwright, M.J., Jordan, M.I.: Graphical models, exponential families, and variational inference. Found. Trends Mach. Learn. 1, 1–305 (2008)
Wang, Z., Zhang, Z., Geng, N.: A message passing algorithm for MRF inference with unknown graphs and its applications. In: Cremers, D., Reid, I., Saito, H., Yang, M.-H. (eds.) ACCV 2014. LNCS, vol. 9006, pp. 288–302. Springer, Cham (2015). https://doi.org/10.1007/978-3-319-16817-3_19
Kappes, J., et al.: A comparative study of modern inference techniques for discrete energy minimization problems. In: CVPR (2013)
Domke, J.: Learning graphical model parameters with approximate marginal inference. TPAMI 35, 2454–2467 (2013)
Taskar, B., Guestrin, C., Koller, D.: Max-Margin Markov Networks. MIT Press, Cambridge (2003)
Tsochantaridis, I., Joachims, T., Hofmann, T., Altun, Y.: Large margin methods for structured and interdependent output variables. JMLR 6, 1453–1484 (2005)
Liu, Z., Li, X., Luo, P., Loy, C.C., Tang, X.: Semantic image segmentation via deep parsing network. In: ICCV (2015)
Lin, G., Shen, C., Hengel, A., Reid, I.: Efficient piecewise training of deep structured models for semantic segmentation. In: CVPR (2016)
Zhang, F., Prisacariu, V., Yang, R., Torr, P.H.: GA-Net: guided aggregation net for end-to-end stereo matching. In: CVPR (2019)
Knobelreiter, P., Reinbacher, C., Shekhovtsov, A., Pock, T.: End-to-end training of hybrid CNN-CRF models for stereo. In: CVPR (2017)
Facciolo, G., Franchis, C., Meinhardt, E.: MGM: a significantly more global matching for stereo vision. In: BMVC (2015)
Hernandez-Juare, D., Chacon, A., Espinosa, A., Vazquez, D., Moure, J., Lopez, A.M.L.: Embedded real-time stereo estimation via semi-global matching on the GPU. In: International Conference on Computational Sciences (2016)
Wainwright, M., Jaakkola, T., Willsky, A.: MAP estimation via agreement on (hyper) trees: message-passing and linear-programming approaches. Trans. Inf. Theory 51(11), 3697–3717 (2005)
Dagum, L., Menon, R.: OpenMP: an industry standard API for shared-memory programming. Comput. Sci. Eng. 5, 46–55 (1998)
Scharstein, D., Szeliski, R.: A taxonomy and evaluation of dense two-frame stereo correspondence algorithms. Int. J. Comput. Vis. 47, 7–42 (2002). https://doi.org/10.1023/A:1014573219977
Scharstein, D., Szeliski, R.: High-accuracy stereo depth maps using structured light. In: CVPR (2003)
Menze, M., Heipke, C., Geiger, A.: Object scene flow. ISPRS J. Photogram. Remote Sens. 140, 60–76 (2018)
Menze, M., Heipke, C., Geiger, A.: Joint 3D estimation of vehicles and scene flow. In: ISPRS Workshop on Image Sequence Analysis (2015)
Schops, T., et al.: A multi-view stereo benchmark with high-resolution images and multi-camera videos. In: CVPR (2017)
Hartley, R., Zisserman, A.: Multiple View Geometry in Computer Vision. Cambridge University Press, Cambridge (2003)
Chen, L.-C., Zhu, Y., Papandreou, G., Schroff, F., Adam, H.: Encoder-decoder with Atrous separable convolution for semantic image segmentation. In: Ferrari, V., Hebert, M., Sminchisescu, C., Weiss, Y. (eds.) ECCV 2018. LNCS, vol. 11211, pp. 833–851. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-01234-2_49
Everingham, M., Eslami, S., Gool, L., Williams, C., Winn, J., Zisserman, A.: The pascal visual object classes challenge a retrospective. Int. J. Comput. Vis. 111, 98–136 (2015). https://doi.org/10.1007/s11263-014-0733-5
Hariharan, B., Arbelaez, P., Bourdev, L., Maji, S., Malik, J.: Semantic contours from inverse detectors. In: ICCV (2011)
Acknowledgement
We would like to thank our colleagues Dylan Campbell and Yao Lu for the discussion of CUDA programming. This work is supported by the Australian Centre for Robotic Vision (CE140100016) and Data61, CSIRO, Canberra, Australia.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
1 Electronic supplementary material
Below is the link to the electronic supplementary material.
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Xu, Z., Ajanthan, T., Hartley, R. (2021). Fast and Differentiable Message Passing on Pairwise Markov Random Fields. In: Ishikawa, H., Liu, CL., Pajdla, T., Shi, J. (eds) Computer Vision – ACCV 2020. ACCV 2020. Lecture Notes in Computer Science(), vol 12624. Springer, Cham. https://doi.org/10.1007/978-3-030-69535-4_32
Download citation
DOI: https://doi.org/10.1007/978-3-030-69535-4_32
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-69534-7
Online ISBN: 978-3-030-69535-4
eBook Packages: Computer ScienceComputer Science (R0)