Research map / 3D vision / Depth and stereo
When Epipolar Constraint Meets Non-Local Operators in Multi-View Stereo
Tianqi Liu, Xinyi Ye, Weiyue Zhao, Zhiyu Pan, Min Shi, Zhiguo Cao
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 |
|---|---|---|---|
| add intra- and cross-epipolar augmentation (IEA and CEA) (ablation) | MVSTER · DTU evaluation set | Overall (mm): 0.318 -> 0.297 | Improved |
| add local augmentation (LA) convolution after IEA and CEA (ablation) | ET · DTU evaluation set | Overall (mm): 0.297 -> 0.291 | Improved |
| add CEA to IEA and LA (ablation) | ET · DTU evaluation set | Overall (mm): 0.294 -> 0.291 | Improved |
| add IEA to CEA and LA (ablation) | ET · DTU evaluation set | Depth error: 5.982 -> 5.754 | Improved |
| use line-to-line search instead of point-to-line search (ablation) | ET · DTU evaluation set | inference time (s): 3.06 -> 0.46 | Improved |
| use line-to-line search instead of point-to-line search (ablation) | ET · DTU evaluation set | Overall (mm): 0.290 -> 0.291 | No clear effect |
| use Epipolar Transformer instead of FMT (ablation) | MVSTER feature encoder · DTU evaluation set | Overall (mm): 0.303 -> 0.291 | Improved |
| use Epipolar Transformer instead of DCN or ASPP (ablation) | MVSTER feature encoder · DTU evaluation set | Overall (mm): 0.309 -> 0.291 (DCN); 0.320 -> 0.291 (ASPP) | Improved |
| embed ET into CasMVSNet feature encoder | CasMVSNet · DTU evaluation set | Overall (mm): 0.355 -> 0.301 | Improved |
| embed ET into CasMVSNet feature encoder | CasMVSNet · Tanks and Temples Advanced | Mean F-score: 31.12 -> 35.65 | Improved |
| use ET-MVSNet versus Gipuma on accuracy | ET-MVSNet · DTU evaluation set | Accuracy (mm): 0.283 (Gipuma) vs 0.329 | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- MVSTER — MVSTER: Epipolar Transformer for Efficient Multi-View Stereo (ECCV 2022) · better on 3 of 3 shared results
- TransMVSNet — TransMVSNet: Global Context-Aware Multi-View Stereo Network With Transformers (CVPR 2022) · better on 2 of 2 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
- MVSTER: Epipolar Transformer for Efficient Multi-View Stereo (ECCV 2022 · similarity 0.91)
- WT-MVSNet: Window-based Transformers for Multi-view Stereo (NeurIPS 2022 · similarity 0.90)
- Epipolar Transformers (CVPR 2020 · similarity 0.88)
- Rethinking Disparity: A Depth Range Free Multi-View Stereo Based on Disparity (AAAI 2023 · similarity 0.88)
- A Global Depth-Range-Free Multi-View Stereo Transformer Network with Pose Embedding (NeurIPS 2024 · similarity 0.87)
- MVSFormer++: Revealing the Devil in Transformer's Details for Multi-View Stereo (ICLR 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.
