Cost Volume Aggregation in Stereo Matching Revisited: A Disparity Classification Perspective
- 11 November 2024
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 33 (10577149) , 6425-6438
Abstract
Cost aggregation plays a critical role in existing stereo matching methods. In this paper, we revisit cost aggregation in stereo matching from disparity classification and propose a generic yet efficient Disparity Context Aggregation (DCA) module to improve the performance of CNN-based methods. Our approach is based on an insight that a coarse disparity class prior is beneficial to disparity regression. To obtain such a prior, we first classify pixels in an image into several disparity classes and treat pixels within the same class as homogeneous regions. We then generate homogeneous region representations and incorporate these representations into the cost volume to suppress irrelevant information while enhancing the matching ability for cost aggregation. With the help of homogeneous region representations, efficient and informative cost aggregation can be achieved with only a shallow 3D CNN. Our DCA module is fully-differentiable and well-compatible with different network architectures, which can be seamlessly plugged into existing networks to improve performance with small additional overheads. It is demonstrated that our DCA module can effectively exploit disparity class priors to improve the performance of cost aggregation. Based on our DCA, we design a highly accurate network named DCANet, which achieves state-of-the-art performance on several benchmarks.Keywords
Funding Information
- National Natural Science Foundation of China (U20A20185, 62372491)
- Basic and Applied Basic Research Foundation of Guangdong Province (2022B1515020103, 2023B1515120087)
- Shenzhen Science and Technology Program (RCYX20200714114641140)
- Shenzhen Key Laboratory of Navigation and Communication Integration (ZDSYS20210623091807023)
- Innohk Initiative, The Government of Hong Kong, SAR (HKSAR), and Laboratory for Artificial Intelligence (AI)-Powered Financial Technologies
This publication has 59 references indexed in Scilit:
- PCW-Net: Pyramid Combination and Warping Cost Volume for Stereo Matching2022
- Context-Enhanced Stereo Transformer2022
- LocalBins: Improving Depth Estimation by Learning Local Distributions2022
- RAFT-Stereo: Multilevel Recurrent Field Transforms for Stereo Matching2021
- RAFT: Recurrent All-Pairs Field Transforms for Optical Flow2020
- Domain-Invariant Stereo Matching Networks2020
- EdgeStereo: An Effective Multi-task Learning Network for Stereo Matching and Edge DetectionInternational Journal of Computer Vision, 2020
- NLCA-Net: a non-local context attention network for stereo matchingAPSIPA Transactions on Signal and Information Processing, 2020
- StereoNet: Guided Hierarchical Refinement for Real-Time Edge-Aware Depth Prediction2018
- High-Resolution Stereo Datasets with Subpixel-Accurate Ground Truth2014