Research map / Label-efficient and robust learning / Domain adaptation and generalization
Domain Generalized Stereo Matching via Hierarchical Visual Transformation
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 hierarchical visual transformations to PSMNet | PSMNet · KITTI 2015, KITTI 2012, Middlebury(H), and ETH3D | EPE: 3.17 -> 1.14; 2.69 -> 0.93; 7.65 -> 1.46; 2.33 -> 0.47 | Helped |
| Add hierarchical visual transformations to GwcNet | GwcNet · KITTI 2015, KITTI 2012, Middlebury(H), and ETH3D | EPE: 3.43 -> 1.15; 2.77 -> 0.88; 7.23 -> 1.29; 2.78 -> 0.46 | Helped |
| Add hierarchical visual transformations to CFNet | CFNet · KITTI 2015, KITTI 2012, Middlebury(H), and ETH3D | EPE: 1.71 -> 1.10; 1.04 -> 0.85; 3.24 -> 1.79; 0.48 -> 0.39 | Helped |
| Add hierarchical visual transformations to RAFT | RAFT · KITTI 2015, KITTI 2012, Middlebury(H), and ETH3D | EPE: 1.26 -> 1.12; 1.01 -> 0.87; 1.92 -> 1.37; 0.36 -> 0.29 | Helped |
| Add Global transformation to PSMNet (ablation) | PSMNet · KITTI 2015 | EPE: 1.64 -> 1.31 | Helped |
| Add Local transformation to PSMNet (ablation) | PSMNet · KITTI 2015 | EPE: 1.64 -> 1.28 | Helped |
| Add Pixel transformation to PSMNet (ablation) | PSMNet · KITTI 2015 | EPE: 1.64 -> 1.27 | Helped |
| Combine Global, Local, and Pixel transformations (ablation) | PSMNet · KITTI 2015 | EPE: 1.14 | Helped |
| Add visual-discrepancy objective (Obj-1) (ablation) | PSMNet · KITTI 2015 | EPE: 1.67 -> 1.31 | Helped |
| Add feature-consistency objective (Obj-2) (ablation) | PSMNet · KITTI 2015 | EPE: 1.67 -> 1.31 | Helped |
| Add visual-discrepancy and feature-consistency objectives together (ablation) | PSMNet · KITTI 2015 | EPE: 1.67 -> 1.14 | Helped |
| Use HVT instead of fine-tuning on KITTI 2015 | PSMNet · DrivingStereo average across weather conditions | D1 (3px): 6.3 -> 5.4 | No clear effect |
Closest papers by meaning
- GraftNet: Towards Domain Generalized Stereo Matching With a Broad-Spectrum and Task-Oriented Feature (CVPR 2022 · similarity 0.92)
- Zoom and Learn: Generalizing Deep Stereo Matching to Novel Domains (CVPR 2018 · similarity 0.91)
- Hierarchical Object-Aware Dual-Level Contrastive Learning for Domain Generalized Stereo Matching (NeurIPS 2024 · similarity 0.91)
- Robust Synthetic-to-Real Transfer for Stereo Matching (CVPR 2024 · similarity 0.90)
- Stereo Risk: A Continuous Modeling Approach to Stereo Matching (ICML 2024 · similarity 0.86)
- FoundationStereo: Zero-Shot 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.