Optics and Precision Engineering, Volume. 32, Issue 3, 445(2024)

Atrous convolution and Bilateral grid network

Jingjing ZHANG1...2,3,*, Xingzhuo DU1,2,3, Shuai ZHI4,5, and Guopeng DING45,* |Show fewer author(s)
Author Affiliations
  • 1School of Automation, China University of Geosciences (Wuhan), Wuhan430074, China
  • 2Hubei Provincial Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems, Wuhan430074, China
  • 3Engineering Research Center of Earth Exploration Intelligent Technology, Ministry of Education, Wuhan40074, China
  • 4Innovation Academy for Microsatellites of Chinese Academy of Sciences, Shanghai201203, China
  • 5Shanghai Microsatellite Engineering Center, Shanghai201203, China
  • show less

    To address the challenges of large-scale and complex network structures in deep learning-based stereo matching, this work introduces a compact yet highly accurate network. The feature extraction module simplifies by removing complex, redundant residual layers and incorporating an Atrous Spatial Pyramid Pooling (ASPP) module to broaden the field of view and enhance contextual information extraction. For cost calculation, three-dimensional (3D) convolutional layers refine stereo matching accuracy through cost aggregation. In addition, a bilateral grid module is integrated into the cost aggregation process, achieving precise disparity maps with reduced resolution demands. Tested on widely-used datasets like KITTI 2015 and Scene Flow, our network demonstrates a significant reduction in parameters by approximately 38% compared to leading networks like Pyramid Stereo Matching Network (PSM-Net), without compromising on experimental accuracy. Notably, it achieves an end-point error (EPE) of 0.86 on the Scene Flow dataset, outperforming many top-performing networks. Thus, our network effectively balances speed and accuracy in stereo matching.

    Keywords
    Tools

    Get Citation

    Copy Citation Text

    Jingjing ZHANG, Xingzhuo DU, Shuai ZHI, Guopeng DING. Atrous convolution and Bilateral grid network[J]. Optics and Precision Engineering, 2024, 32(3): 445

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Save article for my favorites
    Paper Information

    Category:

    Received: Jun. 20, 2023

    Accepted: --

    Published Online: Apr. 2, 2024

    The Author Email: ZHANG Jingjing (dinggp@microsate.com), DING Guopeng (dinggp@microsate.com)

    DOI:10.37188/OPE.20243203.0445

    Topics