Fast Adaptive Depth Estimation Algorithm Based on K-means Segmentation

DOI

10.2991/icmt-13.2013.216How to use a DOI?

Keywords

depth estimation; depth map; image segmentation

Abstract

A fast depth estimation algorithm based on the K-means image segmentation is proposed in this paper. Traditional global-based depth estimation methods using graph-cuts made good performance on the depth map estimation, however, the time cost is a big problem for real time application. A fast segment-based method was discussed in this paper, aiming to reduce time cost and at the same time maintain or even improve the performance of depth map estimation. Firstly, the reference image is segmented by K-means method and then mark each segment as different types. Secondly, apply different matching methods for each kind of segment to get the initial matching cost. Thirdly, correct the depth values of unreliable pixels in each segments. Finally, depth values of each segments are determined by using the final matching cost. The experience results demonstrate the superior performance of the proposed algorithm.

Copyright

© 2013, 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)

TY  - CONF
AU  - Dong Xin
AU  - Wang Guozhong
AU  - Fan Tao
AU  - Li Guoping
AU  - Zhao Haiwu
AU  - Teng Guowei
PY  - 2013/11
DA  - 2013/11
TI  - Fast Adaptive Depth Estimation Algorithm Based on K-means Segmentation
BT  - Proceedings of 3rd International Conference on Multimedia Technology(ICMT-13)
PB  - Atlantis Press
SP  - 1785
EP  - 1792
SN  - 1951-6851
UR  - https://doi.org/10.2991/icmt-13.2013.216
DO  - 10.2991/icmt-13.2013.216
ID  - Xin2013/11
ER  -