Research map / Label-efficient and robust learning / Domain adaptation and generalization
Learning to Adapt for Stereo
Check 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 |
|---|---|---|---|
| Train with L2A+WAd rather than SL+Ad on Carla | Dispnet-Corr1D · KITTI | D1-all (%): 5.26 -> 4.49 | Helped |
| Train with L2A+WAd rather than SL+Ad on Carla | Dispnet-Corr1D · KITTI | EPE: 1.20 -> 1.12 | Helped |
| Train with L2A+WAd rather than SL+Ad on Synthia | Dispnet-Corr1D · KITTI | D1-all (%): 5.33 -> 4.65 | Helped |
| Train with L2A+WAd rather than SL+Ad on Synthia | Dispnet-Corr1D · KITTI | EPE: 1.19 -> 1.14 | Helped |
| Add meta-learning for adaptation instead of supervised SL training on Carla (ablation) | Dispnet-Corr1D · Synthia, unsupervised adaptation | D1-all (%): 25.07 -> 22.69 | Helped |
| Add meta-learning for adaptation instead of supervised SL training on Carla (ablation) | Dispnet-Corr1D · Synthia, unsupervised adaptation | EPE: 3.62 -> 3.08 | Helped |
| Add meta-learning for adaptation instead of supervised SL training on Carla (ablation) | Dispnet-Corr1D · Synthia, supervised adaptation | D1-all (%): 13.89 -> 12.01 | Helped |
| Add meta-learning for adaptation instead of supervised SL training on Carla (ablation) | Dispnet-Corr1D · Synthia, supervised adaptation | EPE: 1.97 -> 1.80 | Helped |
| Add confidence-weighted unsupervised adaptation (ablation) | L2A on Dispnet-Corr1D · Synthia, unsupervised adaptation | D1-all (%): 22.69 -> 21.07 | Helped |
| Add confidence-weighted unsupervised adaptation (ablation) | L2A on Dispnet-Corr1D · Synthia, unsupervised adaptation | EPE: 3.08 -> 2.90 | Helped |
Closest papers by meaning
- RobIA: Robust Instance-aware Continual Test-time Adaptation for Deep Stereo (NeurIPS 2025 · similarity 0.90)
- Federated Online Adaptation for Deep Stereo (CVPR 2024 · similarity 0.87)
- Rectification-specific Supervision and Constrained Estimator for Online Stereo Rectification (CVPR 2025 · similarity 0.87)
- Stereo Risk: A Continuous Modeling Approach to Stereo Matching (ICML 2024 · similarity 0.86)
- Flow-Guided Online Stereo Rectification for Wide Baseline Stereo (CVPR 2024 · similarity 0.85)
- Stereo Anywhere: Robust Zero-Shot Deep Stereo Matching Even Where Either Stereo or Mono Fail (CVPR 2025 · similarity 0.85)
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.