Research map / 3D vision / Depth and stereo
Context-Enhanced Stereo Transformer
Weiyu Guo, Zhaoshuo Li, Yongkui Yang, Zheng Wang, Russell H. Taylor, Mathias Unberath, Alan Yuille, Yingwei Li
Read the paperCheck 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 epipolar self-attention with Axial-Attention (MMP) (ablation) | STTR · Scene Flow test | 3px error: 1.54 -> 1.28 | Improved |
| replace epipolar self-attention with Axial-Attention (MMP) (ablation) | STTR · Middlebury-2014 (zero-shot, Scene Flow trained) | 3px error: 6.93 -> 5.55 | Improved |
| replace epipolar self-attention with Axial-Attention (MMP) (ablation) | STTR · Middlebury-2014 (zero-shot, Scene Flow trained) | EPE: 2.24 -> 2.03 | Improved |
| add Context Enhanced Path M1 (ablation) | STTR + MMP · Scene Flow test | 3px error: 1.28 -> 1.18 | Improved |
| add Context Enhanced Path M2 (ablation) | STTR + MMP · Scene Flow test | 3px error: 1.28 -> 1.20 | Improved |
| add Context Enhanced Path M3 (ablation) | STTR + MMP · Scene Flow test | 3px error: 1.28 -> 1.20 | Improved |
| add Context Enhanced Path M1 (ablation) | STTR + MMP · Middlebury-2014 (zero-shot, Scene Flow trained) | 3px error: 5.55 -> 5.47 | Improved |
| add Context Enhanced Path M2 (ablation) | STTR + MMP · Middlebury-2014 (zero-shot, Scene Flow trained) | 3px error: 5.55 -> 5.38 | Improved |
| add Context Enhanced Path M3 (ablation) | STTR + MMP · Middlebury-2014 (zero-shot, Scene Flow trained) | EPE: 2.03 -> 1.16 | Improved |
| replace epipolar self-attention with Axial-Attention (MMP) (ablation) | STTR · KITTI-2015 hazardous regions (zero-shot) | average EPE: 7.63 -> 6.79 | Improved |
| add Context Enhanced Path M1 (ablation) | STTR + MMP · KITTI-2015 hazardous regions (zero-shot) | average EPE: 6.79 -> 7.71 | Hurt |
| add Context Enhanced Path M2 (ablation) | STTR + MMP · KITTI-2015 hazardous regions (zero-shot) | average EPE: 6.79 -> 7.32 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- PSMNet — PSMNet: Position-Aware Stereo Merging Network for Room Layout Estimation (CVPR 2022) · better on 3 of 3 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Revisiting Stereo Depth Estimation From a Sequence-to-Sequence Perspective With Transformers (ICCV 2021 · similarity 0.89)
- S2M2: Scalable Stereo Matching Model for Reliable Depth Estimation (ICCV 2025 · similarity 0.87)
- Stereo Risk: A Continuous Modeling Approach to Stereo Matching (ICML 2024 · similarity 0.86)
- DEFOM-Stereo: Depth Foundation Model Based Stereo Matching (CVPR 2025 · similarity 0.85)
- Stereo Anywhere: Robust Zero-Shot Deep Stereo Matching Even Where Either Stereo or Mono Fail (CVPR 2025 · similarity 0.85)
- A Global Depth-Range-Free Multi-View Stereo Transformer Network with Pose Embedding (NeurIPS 2024 · 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.
