Proceedings of the 2019 International Conference on Computer, Network, Communication and Information Systems (CNCI 2019)

Improved Weighted Median Filter with Superpixel for Disparity Refinement

Authors
Yuli Fei, Li Cao
Corresponding Author
Yuli Fei
Available Online May 2019.
DOI
10.2991/cnci-19.2019.16How to use a DOI?
Keywords
Stereo Matching, Disparity Refinement, Superpixel Segmentation
Abstract

Stereo matching cannot get high accuracy for disparity estimations, especially at depth boundaries and textureless regions. To solve the problems, a disparity refinement method based on superpixel segmentation is proposed. A weighted median filter with superpixel information is designed. We give a penalty factor for the neighborhood pixels that are not within the same superpixel. Some experiments are done on the Middlebury dataset. The results show that the proposed method can reduce the mismatch rates around occlusion regions and textureless regions, and obtain a highly accurate disparity map.

Copyright
© 2019, the Authors. Published by Atlantis Press.
Open Access
This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

Volume Title
Proceedings of the 2019 International Conference on Computer, Network, Communication and Information Systems (CNCI 2019)
Series
Advances in Computer Science Research
Publication Date
May 2019
ISBN
10.2991/cnci-19.2019.16
ISSN
2352-538X
DOI
10.2991/cnci-19.2019.16How to use a DOI?
Copyright
© 2019, the Authors. Published by Atlantis Press.
Open Access
This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Cite this article

TY  - CONF
AU  - Yuli Fei
AU  - Li Cao
PY  - 2019/05
DA  - 2019/05
TI  - Improved Weighted Median Filter with Superpixel for Disparity Refinement
BT  - Proceedings of the 2019 International Conference on Computer, Network, Communication and Information Systems (CNCI 2019)
PB  - Atlantis Press
SP  - 119
EP  - 124
SN  - 2352-538X
UR  - https://doi.org/10.2991/cnci-19.2019.16
DO  - 10.2991/cnci-19.2019.16
ID  - Fei2019/05
ER  -