Paper
9 September 2022 Loop-tree method for cost aggregation in stereo matching
Author Affiliations +
Proceedings Volume 12328, Second International Conference on Optics and Image Processing (ICOIP 2022); 123281N (2022) https://doi.org/10.1117/12.2644200
Event: Second International Conference on Optics and Image Processing (ICOIP 2022), 2022, Taian, China
Abstract
In high-density stereo correspondence, matching cost aggregation is one of the key links, and the non-local cost aggregation algorithm based on tree structure has attracted people's attention in recent years. On this basis, the clustering algorithm based on non-local tree is studied, and a new cyclic tree structure is given. Each pixel in the image is rooted in eight adjacent tree structures, which we call the first level. We made up the second floor with eight adjacent first floors. Since the algorithm has a natural location in the image pixel structure, it does not need to perform any operations on it. The performance of Middlebury's data set is evaluated, and the results show the application effect of the algorithm proposed in this paper in the current most advanced clustering algorithm.
© (2022) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Qian Zhang and Jun Chen "Loop-tree method for cost aggregation in stereo matching", Proc. SPIE 12328, Second International Conference on Optics and Image Processing (ICOIP 2022), 123281N (9 September 2022); https://doi.org/10.1117/12.2644200
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Image processing algorithms and systems

3D image processing

C++

Reconstruction algorithms

Venus

Aerospace engineering

Error analysis

Back to Top