Research map / Label-efficient and robust learning / Self-supervised and contrastive
Pseudo-Stereo Inputs: A Solution to the Occlusion Challenge in Self-Supervised Stereo Matching
Ruizhi Yang, Xingqiang Li, Jiajun Bai, Jinsong Du
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add pseudo-stereo inputs strategy | PSMNet · SceneFlow | 3-px error (%): 16.85 -> 9.80 | Improved |
| add pseudo-stereo inputs strategy | MonSter · SceneFlow | EPE (px): 5.28 -> 2.32 | Improved |
| add pseudo-stereo inputs strategy | GMStereo · SceneFlow | 3-px error (%): 15.01 -> 8.49 | Improved |
| add pseudo-stereo inputs strategy | PSMNet · Spring (half res) | EPE (px): 2.41 -> 2.30 | Improved |
| add pseudo-stereo inputs strategy | MonSter · Spring (half res) | 3-px error (%): 6.56 -> 4.65 | Improved |
| add pseudo-stereo inputs alone (ablation) | baseline self-supervised stereo · KITTI 2015 train | D1 (%): 5.67 -> 4.99 | Improved |
| add occlusion estimation alone (ablation) | baseline self-supervised stereo · KITTI 2015 train | D1 (%): 5.67 -> 5.73 | Hurt |
| add occlusion estimation to pseudo-stereo inputs (ablation) | Pseudo-Stereo · KITTI 2015 train | D1 (%): 4.99 -> 4.08 | Improved |
| add fully pseudo-stereo inputs strategy to PS and Occ (ablation) | Pseudo-Stereo · KITTI 2015 train | D1 (%): 4.08 -> 4.31 | Hurt |
| add fully pseudo-stereo inputs strategy to PS (ablation) | Pseudo-Stereo · KITTI 2015 train | D1 (%): 4.99 -> 5.17 | Hurt |
| add refined pseudo-image generation (ablation) | Pseudo-Stereo with FPS · KITTI 2015 train | D1 (%): 4.31 -> 4.06 | Improved |
| add refined pseudo-image generation (ablation) | Pseudo-Stereo with FPS · KITTI 2015 train | EPE (px): 1.04 -> 1.01 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Flow2Stereo — Flow2Stereo: Effective Self-Supervised Learning of Optical Flow and Stereo Matching (CVPR 2020) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- DualNet: Robust Self-Supervised Stereo Matching with Pseudo-Label Supervision (AAAI 2025 · similarity 0.90)
- Consistency-aware Self-Training for Iterative-based Stereo Matching (CVPR 2025 · similarity 0.89)
- Any-Stereo: Arbitrary Scale Disparity Estimation for Iterative Stereo Matching (AAAI 2024 · similarity 0.89)
- US$$^3$$Net: Ultralightweight Self-Supervised Stereo Matching Network using Depth-Aware Geometric Soft Occlusion (IJCV 2026 · similarity 0.89)
- On the confidence of stereo matching in a deep-learning era: a quantitative evaluation (TPAMI 2021 · similarity 0.89)
- A Survey on Deep Stereo Matching in the Twenties (IJCV 2025 · similarity 0.89)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
