Research map / Label-efficient and robust learning / Recognition and long-tail
Masked Representation Learning for Domain Generalized Stereo Matching
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 masked representation learning to CFNet | CFNet · KITTI 2012, KITTI 2015, ETH3D, and Middlebury | EPE (mean across epochs): 1.55 -> 1.44; 1.11 -> 1.05; 0.69 -> 0.56; 2.45 -> 1.86 | Helped |
| Add masked representation learning to CFNet | CFNet · KITTI 2012 and ETH3D | threshold error: 5.82 -> 5.03; 7.24 -> 6.63 | Helped |
| Add masked representation learning to LacGwcNet | LacGwcNet · KITTI 2012, KITTI 2015, ETH3D, and Middlebury | EPE (mean across epochs): 1.83 -> 1.43; 1.78 -> 1.41; 2.18 -> 1.00; 2.49 -> 2.40 | Helped |
| Add masked representation learning to LacGwcNet | LacGwcNet · KITTI 2012, KITTI 2015, and Middlebury | threshold error: 9.17 -> 6.57; 8.37 -> 6.08; 18.28 -> 17.30 | Helped |
| Use mask ratio 0.15 rather than 0 (ablation) | CFNet · KITTI 2012 | EPE: 1.11 -> 1.05 | Helped |
| Use mask ratio 0.25 rather than 0 (ablation) | CFNet · KITTI 2012 | EPE: 1.11 -> 1.05 | Helped |
| Use mask ratio 0.35 rather than 0 (ablation) | CFNet · KITTI 2012 | EPE: 1.11 -> 1.13 | Hurt |
| Use mask ratio 0.45 rather than 0 (ablation) | CFNet · KITTI 2012 | EPE: 1.11 -> 1.27 | Hurt |
| Use mask ratio 0.15 rather than 0 (ablation) | LacGwcNet · KITTI 2012 | Out-Noc (%): 1.13 -> 1.15 | Hurt |
| Use mask ratio 0.25 rather than 0 (ablation) | LacGwcNet · KITTI 2012 | Out-Noc (%): 1.13 -> 1.16 | Hurt |
| Use mask ratio 0.35 rather than 0 (ablation) | LacGwcNet · KITTI 2012 | Out-Noc (%): 1.13 -> 1.27 | Hurt |
| Use mask ratio 0.45 rather than 0 (ablation) | LacGwcNet · KITTI 2012 | Out-Noc (%): 1.13 -> 1.39 | Hurt |
Closest papers by meaning
- Domain Generalized Stereo Matching via Hierarchical Visual Transformation (CVPR 2023 · similarity 0.91)
- Zoom and Learn: Generalizing Deep Stereo Matching to Novel Domains (CVPR 2018 · similarity 0.90)
- Hierarchical Object-Aware Dual-Level Contrastive Learning for Domain Generalized Stereo Matching (NeurIPS 2024 · similarity 0.89)
- Robust Synthetic-to-Real Transfer for Stereo Matching (CVPR 2024 · similarity 0.89)
- Stereo Risk: A Continuous Modeling Approach to Stereo Matching (ICML 2024 · similarity 0.88)
- Stereo Anywhere: Robust Zero-Shot Deep Stereo Matching Even Where Either Stereo or Mono Fail (CVPR 2025 · similarity 0.86)
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.