Research map / Label-efficient and robust learning / Recognition and long-tail
StereoGAN: Bridging Synthetic-to-Real Domain Gap by Joint Optimization of Domain Translation and Stereo Matching
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 |
|---|---|---|---|
| jointly train domain translation and stereo matching with stereo constraints | DispNet · Synthia-to-KITTI2015 | D1-all (All, %): 11.14 -> 9.26 | Helped |
| jointly train domain translation and stereo matching with stereo constraints | DispNet · Driving-to-KITTI2015 | D1-all (All, %): 53.07 -> 25.71 | Helped |
| add synthetic feature re-projection loss (ablation) | StereoGAN · Synthia-to-KITTI2015 | D1-all (All, %): 10.02 -> 9.26 | Helped |
| add synthetic feature re-projection loss (ablation) | StereoGAN · Driving-to-KITTI2015 | D1-all (All, %): 26.95 -> 25.71 | Helped |
| add real feature re-projection loss (ablation) | StereoGAN · Synthia-to-KITTI2015 | D1-all (All, %): 9.89 -> 9.26 | Helped |
| add real feature re-projection loss (ablation) | StereoGAN · Driving-to-KITTI2015 | D1-all (All, %): 26.79 -> 25.71 | Helped |
| add correlation consistency loss (ablation) | StereoGAN · Synthia-to-KITTI2015 | D1-all (All, %): 9.45 -> 9.26 | Helped |
| add correlation consistency loss (ablation) | StereoGAN · Driving-to-KITTI2015 | D1-all (All, %): 26.16 -> 25.71 | Helped |
| add mode seeking loss (ablation) | StereoGAN · Synthia-to-KITTI2015 | D1-all (All, %): 9.53 -> 9.26 | Helped |
| add mode seeking loss (ablation) | StereoGAN · Driving-to-KITTI2015 | D1-all (All, %): 25.98 -> 25.71 | Helped |
| use the StereoGAN framework rather than direct synthetic-to-real inference | GwcNet · Driving-to-KITTI2015 | D1-all: 28.21 -> 12.17 | Helped |
| use the StereoGAN framework rather than direct synthetic-to-real inference | GwcNet · Synthia-to-KITTI2015 | D1-all: 7.46 -> 5.74 | Helped |
Closest papers by meaning
- Revisiting Domain Generalized Stereo Matching Networks From a Feature Consistency Perspective (CVPR 2022 · similarity 0.91)
- Robust Synthetic-to-Real Transfer for Stereo Matching (CVPR 2024 · similarity 0.90)
- Domain Generalized Stereo Matching via Hierarchical Visual Transformation (CVPR 2023 · similarity 0.90)
- Hierarchical Object-Aware Dual-Level Contrastive Learning for Domain Generalized Stereo Matching (NeurIPS 2024 · similarity 0.87)
- Stereo Risk: A Continuous Modeling Approach to Stereo Matching (ICML 2024 · similarity 0.87)
- IINet: Implicit Intra-inter Information Fusion for Real-Time Stereo Matching (AAAI 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.