The search functionality is under construction.
The search functionality is under construction.

Asymmetric Learning for Stereo Matching Cost Computation

Zhongjian MA, Dongzhen HUANG, Baoqing LI, Xiaobing YUAN

  • Full Text Views

    0

  • Cite this

Summary :

Current stereo matching methods benefit a lot from the precise stereo estimation with Convolutional Neural Networks (CNNs). Nevertheless, patch-based siamese networks rely on the implicit assumption of constant depth within a window, which does not hold for slanted surfaces. Existing methods for handling slanted patches focus on post-processing. In contrast, we propose a novel module for matching cost networks to overcome this bias. Slanted objects appear horizontally stretched between stereo pairs, suggesting that the feature extraction in the horizontal direction should be different from that in the vertical direction. To tackle this distortion, we utilize asymmetric convolutions in our proposed module. Experimental results show that the proposed module in matching cost networks can achieve higher accuracy with fewer parameters compared to conventional methods.

Publication
IEICE TRANSACTIONS on Information Vol.E103-D No.10 pp.2162-2167
Publication Date
2020/10/01
Publicized
2020/07/13
Online ISSN
1745-1361
DOI
10.1587/transinf.2020EDP7002
Type of Manuscript
PAPER
Category
Artificial Intelligence, Data Mining

Authors

Zhongjian MA
  Chinese Academy of Sciences,University of Chinese Academy of Sciences
Dongzhen HUANG
  Chinese Academy of Sciences
Baoqing LI
  Chinese Academy of Sciences
Xiaobing YUAN
  Chinese Academy of Sciences

Keyword