Segment-Based Stereo Matching Using Belief Propagation and a Self-Adapting Dissimilarity Measure

By Andreas Klaus, Mario Sormann, Konrad F. Karner. International Conference on Pattern Recognition, 2006

994 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/4d0c6d31f263469d4fec4502824c95b9e1082de7

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

Citations in the corpus

Practical Stereo Matching via Cascaded Recurrent Network with Adaptive Correlation (2022, methodology): “Global methods treat stereo matching as an optimization problem, where an explicit cost function is formulated and optimized by belief propagation [20,42,52] or graph cut [4] algorithms.”
High-Accuracy Stereo Matching Based on Adaptive Ground Control Points (2015, methodology): “Sub-precision Results on sphere surface compared with [29].”
Efficient Large-Scale Stereo Matching (2010, methodology): “[14] extend global methods to use mean-shift color segmentation, followed by belief propagation on super-pixels.”
Review of Stereo Vision Algorithms: From Software to Hardware (2008, methodology): “Color segmentation is utilized in (Klaus et al. 2006) as well.”
Displets: Resolving stereo ambiguities using object knowledge (2015, background): “Planes, B-splines and quadratic surfaces have been proposed in [3–5, 24, 69] to better constrain the underlying problems and simplify inference.”
Consistent Depth Maps Recovery from a Video Sequence (2009, background): “G. Zhang and H. Bao are with the State Key Lab of CAD&CG, Zijingang Campus, Zhejiang University, Hangzhou, 310058, P.R. China.”
Fast 3D Scanning with Automatic Motion Compensation (2007, methodology): “The current top performer [10] relies on color segmentation and optimization on the segment level.”
Particle Belief Propagation (2009, methodology): “Belief propagation was applied to the stereo vision problem in Sun et al. (2003), and since then a number of more sophisticated models have also used BP for inference (e.g., Sun et al., 2005; Klaus et al., 2006; Yang et al., 2006).”
Stereo Matching with Color-Weighted Correlation, Hierachical Belief Propagation and Occlusion Handling (2006, methodology): “Chen 22 used K-L divergence, Elidan 18 maintained a priority queue of messages and updated the one with highest residual.”
Efficient Deep Learning for Stereo Matching (2016, background): “They have a long history, dating back to [2] and were shown to be very successful on the Middleburry benchmark [22, 15, 3, 24] as well as on KITTI [25, 26, 27].”
Robust stereo matching with surface normal prediction (2017, methodology): “One representative type of these work are segmentation-based methods, which generally use an image segmentation method to cluster the neighboring pixels with similar colors into the same segment, and then fit the 3D surface for each segment as a 3D plane [23], [33], a Bspline or a quadratic surface ”
Full-reference quality assessment of stereopairs accounting for rivalry (2013, background): “We then focus on the highly relevant phenomenon of binocular rivalry (Section 2.2) and how it affects the quality of distorted stereo images.”
End-to-End Learning of Geometry and Context for Deep Stereo Regression (2017, background): “Global optimization can be accomplished using graph cuts [27] or belief propagation [26], which can be extended to slanted surfaces [6].”
Ieee Transactions on Pattern Analysis and Machine Intelligence 1 Stereo Processing by Semi-global Matching and Mutual Information (methodology): “Since these costs are sensitive to radiometric differences, costs based on image gradients are also used [3].”
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.”
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].”
On building an accurate stereo matching system on graphics hardware (2011, methodology): “Although many algorithms are introduced every year, the two concerns tend to be contradictory in the reported results: accurate stereo methods are usually time consuming [6, 17, 20], while GPU-based methods achieve high processing speed with relatively low disparity precision [10, 18, 24].”