Research map / 3D vision / Depth and stereo
Digging Into Uncertainty in Self-Supervised Multi-View Stereo
Hongbin Xu, Zhipeng Zhou, Yali Wang, Wenxiong Kang, Baigui Sun, Hao Li, Yu Qiao
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 flow-depth consistency loss L_fc (ablation) | U-MVS (MVSNet) · DTU evaluation | overall error: 0.5436 -> 0.4820 | Improved |
| add uncertainty-aware self-training loss L_uc (ablation) | U-MVS (MVSNet) · DTU evaluation | overall error: 0.4820 -> 0.4501 | Improved |
| add flow-depth consistency loss L_fc (ablation) | U-MVS (CascadeMVSNet) · DTU evaluation | overall error: 0.4041 -> 0.3789 | Improved |
| add uncertainty-aware self-training loss L_uc (ablation) | U-MVS (CascadeMVSNet) · DTU evaluation | overall error: 0.3789 -> 0.3537 | Improved |
| add flow-depth consistency loss L_fc (ablation) | U-MVS (CascadeMVSNet) · DTU evaluation | completeness error: 0.3641 -> 0.3833 | Hurt |
| add flow-depth consistency loss L_fc (ablation) | U-MVS (CascadeMVSNet) · DTU evaluation | accuracy error: 0.4442 -> 0.3745 | Improved |
| add L_fc and L_uc to photometric loss (ablation) | U-MVS (MVSNet) · DTU evaluation | accuracy error: 0.5527 -> 0.4695 | Improved |
| self-supervised U-MVS instead of supervised training with same CascadeMVSNet backbone | CascadeMVSNet · Tanks and Temples intermediate | mean F-score: 56.42 -> 57.15 | Improved |
| self-supervised U-MVS instead of supervised training with same CascadeMVSNet backbone | CascadeMVSNet · Tanks and Temples advanced | mean F-score: 31.12 -> 30.97 | No clear effect |
| self-supervised U-MVS instead of supervised training with same CascadeMVSNet backbone | CascadeMVSNet · DTU evaluation | overall error: 0.355 -> 0.3537 | No clear effect |
| self-supervised U-MVS instead of supervised training with same CascadeMVSNet backbone | CascadeMVSNet · DTU evaluation | accuracy error: 0.325 -> 0.354 | Hurt |
| self-supervised U-MVS instead of supervised training with same CascadeMVSNet backbone | CascadeMVSNet · DTU evaluation | completeness error: 0.385 -> 0.3535 | Improved |
Closest papers by meaning
- Self-Supervised Learning of Depth Inference for Multi-View Stereo (CVPR 2021 · similarity 0.90)
- KD-MVS: Knowledge Distillation Based Self-Supervised Learning for Multi-View Stereo (ECCV 2022 · similarity 0.90)
- On the Uncertainty of Self-Supervised Monocular Depth Estimation (CVPR 2020 · similarity 0.90)
- Just a Few Points Are All You Need for Multi-View Stereo: A Novel Semi-Supervised Learning Method for Multi-View Stereo (ICCV 2021 · similarity 0.89)
- DualNet: Robust Self-Supervised Stereo Matching with Pseudo-Label Supervision (AAAI 2025 · similarity 0.87)
- Consistency-aware Self-Training for Iterative-based 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.
