Research map / Label-efficient and robust learning / Recognition and long-tail
Degradation-Agnostic Correspondence From Resolution-Asymmetric Stereo
Xihao Chen, Zhiwei Xiong, Zhen Cheng, Jiayong Peng, Yueyi Zhang, Zheng-Jun Zha
Check 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 |
|---|---|---|---|
| replace photometric consistency with feature-metric consistency from the stereo network's own feature extractor | PSMNet unsupervised stereo matching · Inria_SLFD, Middlebury, HCI, and KITTI2015 (BIC, s=4) | 3PE (%): Inria_SLFD: 8.89 -> 7.70; Middlebury: 6.76 -> 5.78; HCI: 5.13 -> 4.08; KITTI2015: 9.94 -> 8.66 | Improved |
| add self-boosting stages that update the feature-metric loss progressively (ablation) | PSMNet on Inria_SLFD · s=4 | 3PE (%): 12.56 -> 9.22 -> 7.80 -> 7.70 | Improved |
| add self-boosting stages that update the feature-metric loss progressively (ablation) | PSMNet on Inria_SLFD · s=6 | 3PE (%): 21.47 -> 13.92 -> 10.54 -> 9.88 | Improved |
| add self-boosting stages that update the feature-metric loss progressively (ablation) | PSMNet on Inria_SLFD · s=8 | 3PE (%): 27.93 -> 18.47 -> 14.30 -> 13.30 | Improved |
| use a symmetric photometric loss instead of an asymmetric photometric loss (ablation) | PSMNet on Inria_SLFD · s=2, 4, 6, 8 | 3PE (%): 7.28 -> 6.38; 12.56 -> 6.39; 22.72 -> 6.58; 27.93 -> 7.52 | Improved |
| use symmetric rather than asymmetric inputs to the feature extractor while keeping the asymmetric loss (ablation) | PSMNet on Inria_SLFD · s=2, 4, 6, 8 | 3PE (%): 7.28 -> 7.22; 12.56 -> 10.01; 22.72 -> 16.31; 27.93 -> 21.93 | Improved |
| use CL features as the feature-metric space (ablation) | BaseNet · simulated asymmetric stereo datasets | 3PE (%): Inria_SLFD BIC: 12.56 -> 12.97; HCI BIC: 5.95 -> 7.80; KITTI2015 BIC: 11.32 -> 17.04 | Hurt |
| use autoencoder features as the feature-metric space (ablation) | BaseNet · simulated asymmetric stereo datasets | 3PE (%): Inria_SLFD BIC: 12.56 -> 10.47; HCI BIC: 5.95 -> 5.13; KITTI2015 BIC: 11.32 -> 10.53 | Mixed |
| replace PSMNet with iResNet while retaining the proposed feature-metric method (ablation) | Ours on simulated datasets · BIC, s=4 | 3PE (%): Inria_SLFD: 18.80 -> 9.83; HCI: 18.58 -> 5.83; Middlebury: 10.92 -> 8.39; KITTI2015: 17.82 -> 10.86 | Improved |
| use supervised disparity labels instead of unsupervised learning (ablation) | BaseNet on Middlebury-trained models · KITTI2015 test, IG, s=4 | 3PE (%): 13.14 -> 19.46 | Hurt |
| use RCAN super-resolution preprocessing | BaseNet · real-world Huawei P30 tele-wide stereo | visual disparity quality | No clear effect |
Closest papers by meaning
- Unsupervised Deep Asymmetric Stereo Matching With Spatially-Adaptive Self-Similarity (CVPR 2023 · similarity 0.89)
- Revisiting Domain Generalized Stereo Matching Networks From a Feature Consistency Perspective (CVPR 2022 · similarity 0.88)
- Consistency-aware Self-Training for Iterative-based Stereo Matching (CVPR 2025 · similarity 0.88)
- DualNet: Robust Self-Supervised Stereo Matching with Pseudo-Label Supervision (AAAI 2025 · similarity 0.86)
- Any-Stereo: Arbitrary Scale Disparity Estimation for Iterative Stereo Matching (AAAI 2024 · similarity 0.86)
- Stereo Risk: A Continuous Modeling Approach to Stereo Matching (ICML 2024 · 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.
