MoCha-Stereo: Motif Channel Attention Network for Stereo Matching
This addresses edge detail mismatches in stereo matching for applications like autonomous driving, but it is incremental as it builds on existing learning-based techniques.
The paper tackles the problem of losing geometrical structure information in stereo matching, which causes edge detail mismatches, by proposing MoCha-Stereo with Motif Channel Correlation Volume and Reconstruction Error Motif Penalty, achieving 1st rank on KITTI-2015 and KITTI-2012 Reflective leaderboards.
Learning-based stereo matching techniques have made significant progress. However, existing methods inevitably lose geometrical structure information during the feature channel generation process, resulting in edge detail mismatches. In this paper, the Motif Cha}nnel Attention Stereo Matching Network (MoCha-Stereo) is designed to address this problem. We provide the Motif Channel Correlation Volume (MCCV) to determine more accurate edge matching costs. MCCV is achieved by projecting motif channels, which capture common geometric structures in feature channels, onto feature maps and cost volumes. In addition, edge variations in %potential feature channels of the reconstruction error map also affect details matching, we propose the Reconstruction Error Motif Penalty (REMP) module to further refine the full-resolution disparity estimation. REMP integrates the frequency information of typical channel features from the reconstruction error. MoCha-Stereo ranks 1st on the KITTI-2015 and KITTI-2012 Reflective leaderboards. Our structure also shows excellent performance in Multi-View Stereo. Code is avaliable at https://github.com/ZYangChen/MoCha-Stereo.