Spatiotemporal Stereo Matching with 3D Disparity Profiles

Yongho Shin and Kuk-Jin Yoon

Abstract

Adaptive support weights and over-parameterized disparity estimation truly improve the accuracy of stereo matching by enabling window-based similarity measures to handle depth discontinuities and non-fronto-parallel surfaces more effectively. Nevertheless, a disparity map sequence obtained in a frame-by-frame manner still tends to be inconsistent even with the use of state-of-the-art stereo matching methods. To solve this inconsistency problem, we propose a window-based spatiotemporal stereo matching method. We exploit the 3D disparity profile, which represents the disparities and window normals over multiple frames, and incorporate it into the PatchMatch Belief Propagation (PMBP) framework. Here, to make the 3D disparity profile more reliable, we also present the optical flow transfer method. Experimental results show the proposed method yields more consistent disparity map sequences than does the original PMBP-based method.

Session

Poster 2

Files

Extended Abstract (PDF, 4M)
Paper (PDF, 9M)
Supplemental Materials (ZIP, 29M)

DOI

10.5244/C.29.152
https://dx.doi.org/10.5244/C.29.152

Citation

Bibtex

@inproceedings{BMVC2015_152,
	title={Spatiotemporal Stereo Matching with 3D Disparity Profiles},
	author={Yongho Shin and Kuk-Jin Yoon},
	year={2015},
	month={September},
	pages={152.1-152.12},
	articleno={152},
	numpages={12},
	booktitle={Proceedings of the British Machine Vision Conference (BMVC)},
	publisher={BMVA Press},
	editor={Xianghua Xie, Mark W. Jones, and Gary K. L. Tam},
	doi={10.5244/C.29.152},
	isbn={1-901725-53-7},
	url={https://dx.doi.org/10.5244/C.29.152}
}