Stereo Matching Using Belief Propagation

By Harry Shum, Nanning Zheng. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002

1,637 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Media Technology

View Paper: https://www.semanticscholar.org/paper/b81293c99535d5faff42c15e61bf52fbc6a36bb5

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Media Technology

Citations in the corpus

Cascade Residual Learning: A Two-Stage Convolutional Neural Network for Stereo Matching (2017, background): “, [11, 26], consists of four steps [24]: 1) matching cost computation; 2) cost aggregation; 3) disparity optimization (derive the disparity from the cost volume); and 4) disparity refinement (post-process the disparity).”
Stereo correspondence by dynamic programming on a tree (2005, methodology): “Recently, graph-cuts [6, 15] and belief propagation methods [20, 9] have been applied quite successfully to optimize equation (1).”
Residual Splash for Optimally Parallelizing Belief Propagation (2009, methodology): “Nonetheless, BP on “loopy” graphs (often referred to as Loopy BP ) is used extensively with great success as an approximate inference algorithm [McEliece et al., 1998, Sun et al., 2003, Yedidia et al., 2003, Yanover and Weiss, 2002, Yanover et al., 2007].”
Cascade Cost Volume for High-Resolution Multi-View Stereo and Stereo Matching (2019, methodology): “More specifically, works in [16, 35] use belief propagation and semi-global matching [10] to approximate the global optimization with dynamic programming.”
A non-local cost aggregation method for stereo matching (2012, methodology): “Popular global methods include dynamic programming [2, 23], belief propagation [13, 14, 15, 16] and graph cuts [3].”
AANet: Adaptive Aggregation Network for Efficient Stereo Matching (2020, methodology): “Global methods usually solve an optimization problem by minimizing a global objective function that contains data and smoothness terms [31, 17], while local methods only consider neighbor information [40, 12], making themselves much faster than global methods [23, 29].”
Image completion with structure propagation (2005, background): “Recent empirical results on several computer vision problems [Freeman et al. 2000; Sun et al. 2002] show that belief propagation is often a very good approximation even for graphs with thousands of loops.”
Distributed message passing for large scale graphical models (2011, background): “Here we focus on message passing algorithms which have proven very effective to solve vision tasks such as super-resolution [3], stereo reconstruction [26], segmentation [15], denoising [13], and inpainting [16].”
Stereo for Image-Based Rendering using Image Over-Segmentation (2007, methodology): “The values at each node, called the disparity space image (DSI) (Intille and Bobick, 1994), can then be updated using loopy belief propagation (Sun et al., 2003).”
Consistent Depth Maps Recovery from a Video Sequence (2009, background): “The reconstructed depths usually serve as a valuable source of information, and facilitate applications in various fields, including 3D modeling, layer separation, image-based rendering, and video editing.”
Fields of Experts (2009, background): “Markov random fields have found many areas of application including image denoising (Sebastiani and Godtliebsen, 1997), stereo (Sun et al., 2003), optical flow estimation (Heitz and Bouthemy, 1993), texture classification (Varma and Zisserman, 2005), to name a few.”
Convergent Tree-Reweighted Message Passing for Energy Minimization (2006, methodology): “If a function falls outside this class then one has to use other techniques such as max-product belief propagation (BP) [6,22,27,35].”
A Taxonomy and Evaluation of Dense Two-Frame Stereo Correspondence Algorithms (2001, methodology): “[109] (a MRF formulation using Bayesian belief propagation);”
Linear Programming Relaxations and Belief Propagation - An Empirical Study (2006, background): “1999; Kolmogorov and Zabih, 2004) and Belief Propagation (Felzenszwalb and Huttenlocher, 2004; Tappen and Freeman, 2003; Sun et al., 2002).”
Soft 3D reconstruction for view synthesis (2017, methodology): “Global optimization methods such as [Felzenszwalb and Huttenlocher 2006; Kolmogorov and Zabih 2001; Sun et al. 2003] seek to minimize a global cost function that also optimizes for depth smoothness, while ‘local’ methods attempt to achieve the same more efficiently (or even at real-time rates) using”
Group-Wise Correlation Stereo Network (2019, methodology): “In contrast, global methods minimize a target function to solve the optimal disparity map, which usually takes both matching costs and smoothness priors into consideration, such as belief propagation [23, 9] and graph cut [11].”
Beyond pixels: exploring new representations and applications for motion analysis (2009, positive): “When the feature values are different, or other information such as occlusion needs to be taken into account, one can use belief propagation [40, 110] and graph cuts [16, 56] to optimize objective functions formulated on Markov random fields.”