Semi-Supervised Disparity Estimation with Deep Feature Reconstruction
Not on the map. Why?
Abstract (arXiv)
Despite the success of deep learning in disparity estimation, the domain generalization gap remains an issue. We propose a semi-supervised pipeline that successfully adapts DispNet to a real-world domain by joint supervised training on labeled synthetic data and self-supervised training on unlabeled real data. Furthermore, accounting for the limitations of the widely-used photometric loss, we analyze the impact of deep feature reconstruction as a promising supervisory signal for disparity estimation.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 13 of its 21 references that are in the map. First posted on arXiv 1 Jun 2021.
Builds on
13 of 21 references.
References · 13
- 2015A Large Dataset to Train Convolutional Networks for Disparity, Optical Flow, and Scene Flow Estimationcited by 878
- 2017End-to-End Learning of Geometry and Context for Deep Stereo Regressioncited by 359
- 2019Group-wise Correlation Stereo Networkcited by 235
- 2020Hierarchical Neural Architecture Search for Deep Stereo Matchingcited by 166
- 2018Unsupervised Learning of Monocular Depth Estimation and Visual Odometry with Deep Feature Reconstructioncited by 159
- 2020Feature-metric Loss for Self-supervised Learning of Depth and Egomotioncited by 140
- 2019Hierarchical Deep Stereo Matching on High-resolution Imagescited by 126
- 2020What Matters in Unsupervised Optical Flowcited by 124
- 2020HITNet: Hierarchical Iterative Tile Refinement Network for Real-time Stereo Matchingcited by 120
- 2018SegStereo: Exploiting Semantic Information for Disparity Estimationcited by 88
- 2018Occlusions, Motion and Depth Boundaries with a Generic Network for Disparity, Optical Flow or Scene Flow Estimationcited by 69
- 2020Reversing the cycle: self-supervised deep stereo through enhanced monocular distillationcited by 34
Built on by
Citing papers per year: 2026: 1.
1 paper.