Research map / 3D vision / Depth and stereo
P-MVSNet: Learning Patch-Wise Matching Confidence Aggregation for Multi-View Stereo
Keyang Luo, Tao Guan, Lili Ju, Haipeng Huang, Yawei Luo
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 patch-wise confidence aggregation (ablation) | P-MVSNet (D_l2) · DTU eval (depth maps) | mean absolute depth error: 5.54 -> 5.26 | Improved |
| add patch-wise confidence aggregation (ablation) | P-MVSNet (D_l1) · DTU eval (depth maps) | mean absolute depth error: 5.74 -> 5.43 | Improved |
| use hybrid 3D U-Net instead of generic 3D U-Net (ablation) | P-MVSNet (D_l2) · DTU eval (depth maps) | mean absolute depth error: 5.82 -> 5.26 | Improved |
| use hybrid 3D U-Net instead of generic 3D U-Net (ablation) | P-MVSNet (D_l1) · DTU eval (depth maps) | mean absolute depth error: 6.13 -> 5.43 | Improved |
| add patch-wise confidence aggregation (ablation) | P-MVSNet (D_l2) · DTU eval (depth maps) | prediction precision (3 sigma): 89.25% -> 90.88% | Improved |
| add patch-wise confidence aggregation (ablation) | P-MVSNet (D_l1) · DTU eval (depth maps) | prediction precision (sigma): 73.06% -> 73.97% | Improved |
| use hybrid 3D U-Net instead of generic 3D U-Net (ablation) | P-MVSNet (D_l1) · DTU eval (depth maps) | prediction precision (3 sigma): 87.21% -> 88.47% | Improved |
| use hybrid 3D U-Net instead of generic 3D U-Net (ablation) | P-MVSNet (D_l2) · DTU eval (depth maps) | prediction precision (sigma): 73.66% -> 75.43% | Improved |
| use L2 depth regression instead of L1 for depth maps | P-MVSNet · DTU eval (depth maps) | mean absolute depth error: 5.43 -> 5.26 | Improved |
| replace MVSNet with P-MVSNet | MVSNet · DTU eval (depth maps) | mean absolute depth error: 7.25 -> 5.26 | Improved |
| replace MVSNet with P-MVSNet | MVSNet · DTU eval (fused point clouds) | overall error: 0.462 -> 0.420 | Improved |
| replace Gipuma with P-MVSNet (accuracy) | Gipuma · DTU eval (fused point clouds) | mean accuracy error: 0.274 -> 0.406 | Hurt |
Later papers that compared with this method
5 later accepted papers report P-MVSNet as a baseline in their results tables; 4 of them report a better number on most of the results they share with it.
- Pyramid Multi-view Stereo Net with Self-adaptive View Aggregation (ECCV 2020 · better on 3 of 5 shared results)
- Dense Hybrid Recurrent Multi-view Stereo Net with Dynamic Consistency Checking (ECCV 2020 · better on 2 of 2 shared results)
- Attention-Aware Multi-View Stereo (CVPR 2020 · better on 1 of 1 shared result)
- Cascade Cost Volume for High-Resolution Multi-View Stereo and Stereo Matching (CVPR 2020 · better on 1 of 1 shared result)
- Cost Volume Pyramid Based Depth Inference for Multi-View Stereo (CVPR 2020 · better on 0 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
- 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 0.91)
- Fast-MVSNet: Sparse-to-Dense Multi-View Stereo With Learned Propagation and Gauss-Newton Refinement (CVPR 2020 · similarity 0.91)
- MVS2D: Efficient Multi-View Stereo via Attention-Driven 2D Convolutions (CVPR 2022 · similarity 0.91)
- A Confidence-Based Iterative Solver of Depths and Surface Normals for Deep Multi-View Stereo (ICCV 2021 · similarity 0.91)
- AANet: Adaptive Aggregation Network for Efficient Stereo Matching (CVPR 2020 · similarity 0.91)
- Generalized Binary Search Network for Highly-Efficient Multi-View Stereo (CVPR 2022 · similarity 0.90)
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.
