Research map / 3D vision / Depth and stereo
MVSCRF: Learning Multi-View Stereo With Conditional Random Fields
Youze Xue, Jiansheng Chen, Weitao Wan, Yiqing Huang, Cheng Yu, Tianpeng Li, Jiayu Bao
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 U-shape feature extractor (ablation) | MVSNet · DTU | accuracy distance (mm): 0.396 -> 0.351 | Improved |
| add U-shape feature extractor (ablation) | MVSNet · DTU | overall mean distance (mm): 0.462 -> 0.432 | Improved |
| add U-shape feature extractor (ablation) | MVSNet · DTU | completeness distance (mm): 0.527 -> 0.513 | Improved |
| add multi-scale CRFs module on top of U-extractor (ablation) | MVSNet + U-extractor · DTU | completeness distance (mm): 0.513 -> 0.426 | Improved |
| add multi-scale CRFs module on top of U-extractor (ablation) | MVSNet + U-extractor · DTU | overall mean distance (mm): 0.432 -> 0.398 | Improved |
| add multi-scale CRFs module on top of U-extractor (ablation) | MVSNet + U-extractor · DTU | accuracy distance (mm): 0.351 -> 0.371 | Hurt |
| use multi-scale CRF instead of single-scale CRF (ablation) | MVSCRF · Tanks and Temples | mean f-score: 44.00 -> 45.73 | Improved |
| use multi-scale CRF instead of single-scale CRF (ablation) | MVSCRF · DTU (<1mm) | f-score (%): 79.77 -> 80.02 | No clear effect |
| add MVSCRF components instead of denser depth sampling (256 vs 512 samples) (ablation) | R-MVSNet · DTU | overall mean distance (mm): 0.417 -> 0.398 | Improved |
| add post-processing and fusibile fusion to reproduced MVSNet | MVSNet · DTU | mean distance (mm): 0.462 -> 0.462 | No clear effect |
Later papers that compared with this method
1 later accepted paper reports MVSCRF as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- Cost Volume Pyramid Based Depth Inference for Multi-View Stereo (CVPR 2020 · better on 1 of 1 shared result)
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
- MVSNet: Depth Inference for Unstructured Multi-view Stereo (ECCV 2018 · similarity 0.91)
- 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 0.90)
- Fast-MVSNet: Sparse-to-Dense Multi-View Stereo With Learned Propagation and Gauss-Newton Refinement (CVPR 2020 · similarity 0.90)
- MVS2D: Efficient Multi-View Stereo via Attention-Driven 2D Convolutions (CVPR 2022 · similarity 0.89)
- RayMVSNet: Learning Ray-Based 1D Implicit Fields for Accurate Multi-View Stereo (CVPR 2022 · similarity 0.89)
- A Confidence-Based Iterative Solver of Depths and Surface Normals for Deep Multi-View Stereo (ICCV 2021 · similarity 0.89)
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.
