Research map / 3D vision / Depth and stereo
Left-Right Comparative Recurrent Model for 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 |
|---|---|---|---|
| Use LRCR instead of CRL | Stereo matching · KITTI 2015 testing set | ALL error rate: 2.67 -> 3.03 | Hurt |
| Use LRCR instead of GC-NET | Stereo matching · KITTI 2015 testing set | ALL error rate: 2.87 -> 3.03 | Hurt |
| Use LRCR instead of GC-NET | Stereo matching · Scene Flow testing set | EPE: 2.51 -> 2.02 | Helped |
| Use LRCR instead of MC-CNN-acrt | Stereo matching · Middlebury testing set | Error Rate: 8.08 -> 7.43 | Helped |
| Add recurrent refinement from t=1 to t=5 | LRCR · KITTI 2015 validation set | EPE: 0.971 -> 0.806 | Helped |
| Add recurrent refinement from t=1 to t=5 | LRCR · Scene Flow testing set | EPE: 6.33 -> 2.02 | Helped |
| Add left-right comparison at recurrent steps (ablation) | LRCR · KITTI 2015 validation set | EPE: 0.913 -> 0.806 | Helped |
| Add learned attention instead of disparity-map subtraction (ablation) | LRCR · KITTI 2015 validation set | EPE: 0.825 -> 0.806 | Helped |
| Use recurrent refinement instead of the non-recurrent model (ablation) | LRCR · KITTI 2015 validation set | EPE: 0.939 -> 0.806 | Helped |
| Use recurrent refinement instead of the non-recurrent model (ablation) | LRCR · Scene Flow testing set | EPE: 5.26 -> 2.02 | Helped |
| Use the first recurrent step rather than the non-recurrent model (ablation) | LRCR · Scene Flow testing set | EPE: 5.26 -> 6.33 | Hurt |
Closest papers by meaning
- Modeling Stereo-Confidence out of the End-to-End Stereo-Matching Network via Disparity Plane Sweep (AAAI 2024 · similarity 0.89)
- Any-Stereo: Arbitrary Scale Disparity Estimation for Iterative Stereo Matching (AAAI 2024 · similarity 0.87)
- Stereo Anywhere: Robust Zero-Shot Deep Stereo Matching Even Where Either Stereo or Mono Fail (CVPR 2025 · similarity 0.86)
- Rectification-specific Supervision and Constrained Estimator for Online Stereo Rectification (CVPR 2025 · similarity 0.86)
- DEFOM-Stereo: Depth Foundation Model Based Stereo Matching (CVPR 2025 · similarity 0.86)
- Consistency-aware Self-Training for Iterative-based Stereo Matching (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.