Hierarchical Belief Propagation on Image Segmentation Pyramid

被引:2
作者
Yan, Tingman [1 ]
Yang, Xilian [1 ]
Yang, Genke [2 ,3 ]
Zhao, Qunfei [2 ,3 ]
机构
[1] Shanghai Jiao Tong Univ, Inst Image Proc & Pattern Recognit, Dept Automat, Shanghai 200240, Peoples R China
[2] Ningbo Artificial Intelligence Inst, Ningbo 315012, Peoples R China
[3] Shanghai Jiao Tong Univ, Dept Automat, Shanghai 200240, Peoples R China
基金
中国国家自然科学基金;
关键词
Image segmentation; Inference algorithms; Labeling; Belief propagation; Three-dimensional displays; Costs; Computational modeling; Markov random field; belief propagation; stereo matching;
D O I
10.1109/TIP.2023.3299192
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
The Markov random field (MRF) for stereo matching can be solved using belief propagation (BP). However, the solution space grows significantly with the introduction of high-resolution stereo images and 3D plane labels, making the traditional BP algorithms impractical in inference time and convergence. We present an accurate and efficient hierarchical BP framework using the representation of the image segmentation pyramid (ISP). The pixel-level MRF can be solved by a top-down inference on the ISP. We design a hierarchy of MRF networks using the graph of superpixels at each ISP level. From the highest/image to the lowest/pixel level, the MRF models can be efficiently inferred with constant global guidance using the optimal labels of the previous level. The large texture-less regions can be handled effectively by the MRF model on a high level. The advanced 3D continuous labels and a novel support-points regularization are integrated into our framework for stereo matching. We provide a data-level parallelism implementation which is orders of magnitude faster than the best graph cuts (GC) algorithm. The proposed framework, HBP-ISP, outperforms the best GC algorithm on the Middlebury stereo matching benchmark.
引用
收藏
页码:4432 / 4442
页数:11
相关论文
共 50 条
  • [31] Belief-propagation on edge images for stereo analysis of image sequences
    Guan, Shushi
    Klette, Reinhard
    [J]. ROBOT VISION, PROCEEDINGS, 2008, 4931 : 291 - 302
  • [32] SCoBeP: Dense image registration using sparse coding and belief propagation
    Barzigar, Nafise
    Roozgard, Aminmohammad
    Cheng, Samuel
    Verma, Pramode
    [J]. JOURNAL OF VISUAL COMMUNICATION AND IMAGE REPRESENTATION, 2013, 24 (02) : 137 - 147
  • [33] IMPROVED SIFT-BASED IMAGE REGISTRATION USING BELIEF PROPAGATION
    Cheng, Samuel
    Stankovic, Vladimir
    Stankovic, Lina
    [J]. 2009 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, VOLS 1- 8, PROCEEDINGS, 2009, : 2909 - +
  • [34] Hardware-Efficient Belief Propagation
    Liang, Chia-Kai
    Cheng, Chao-Chung
    Lai, Yen-Chieh
    Chen, Liang-Gee
    Chen, Homer H.
    [J]. IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY, 2011, 21 (05) : 525 - 537
  • [35] Automatic image segmentation by wave propagation
    Porikli, F
    [J]. IMAGE PROCESSING: ALGORITHMS AND SYSTEMS III, 2004, 5298 : 536 - 543
  • [36] CPFNet: Context Pyramid Fusion Network for Medical Image Segmentation
    Feng, Shuanglang
    Zhao, Heming
    Shi, Fei
    Cheng, Xuena
    Wang, Meng
    Ma, Yuhui
    Xiang, Dehui
    Zhu, Weifang
    Chen, Xinjian
    [J]. IEEE TRANSACTIONS ON MEDICAL IMAGING, 2020, 39 (10) : 3008 - 3018
  • [37] Bayesian Image Modeling by Means of a Generalized Sparse Prior and Loopy Belief Propagation
    Tanaka, Kazuyuki
    Yasuda, Muneki
    Titterington, D. Michael
    [J]. JOURNAL OF THE PHYSICAL SOCIETY OF JAPAN, 2012, 81 (11)
  • [38] New UAV image registration method based on geometric constrained belief propagation
    Shuangming Zhao
    Guorong Yu
    Yunfan Cui
    [J]. Multimedia Tools and Applications, 2018, 77 : 24143 - 24163
  • [39] New UAV image registration method based on geometric constrained belief propagation
    Zhao, Shuangming
    Yu, Guorong
    Cui, Yunfan
    [J]. MULTIMEDIA TOOLS AND APPLICATIONS, 2018, 77 (18) : 24143 - 24163
  • [40] Neural Belief Propagation for Scene Graph Generation
    Liu, Daqi
    Bober, Miroslaw
    Kittler, Josef
    [J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2023, 45 (08) : 10161 - 10172