Research map / 3D vision / Depth and stereo
MVPSNet: Fast Generalizable Multi-view Photometric Stereo
Dongxu Zhao, Daniel Lichy, Pierre-Nicolas Perrin, Jan-Michael Frahm, Soumyadip Sengupta
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 |
|---|---|---|---|
| retrain CasMVSNet on synthetic sMVPS data instead of DTU (ablation) | CasMVSNet · DiLiGenT-MV (after ICP) | L1 Chamfer distance: 1.83 -> 1.24 | Improved |
| add multiple lights per view (3 instead of 1) through LAFM (ablation) | MVPSNet · DiLiGenT-MV (after ICP) | L1 Chamfer distance: 1.09 -> 0.82 | Improved |
| use LAFM aggregation versus single-light MVS features (ablation) | MVPSNet · DiLiGenT-MV (after ICP) | L1 Chamfer distance: 1.24 -> 1.09 | Improved |
| add surface normal loss (ablation) | MVPSNet · DiLiGenT-MV (after ICP) | L1 Chamfer distance: 0.79 -> 0.82 | No clear effect |
| add surface normal loss (ablation) | MVPSNet · DiLiGenT-MV (after ICP) | F-score: 0.985 -> 0.985 | No clear effect |
| add multi-light MVPS network versus light-agnostic MVS | MVPSNet vs TransMVSNet-RT · DiLiGenT-MV (after ICP) | L1 Chamfer distance: 1.09 -> 0.82 | Improved |
| use transformer MVS (TransMVSNet) instead of CasMVSNet when retrained on sMVPS | MVS baselines · DiLiGenT-MV (after ICP) | L1 Chamfer distance: 1.24 -> 1.09 | Improved |
| skip ICP registration | MVPSNet · DiLiGenT-MV | L1 Chamfer distance: 0.82 -> 0.91 | Hurt |
| use PS-NeRF per-scene optimization versus MVPSNet | PS-NeRF · DiLiGenT-MV (after ICP) | L1 Chamfer distance: 0.81 -> 0.82 | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- TransMVSNet — TransMVSNet: Global Context-Aware Multi-View Stereo Network With Transformers (CVPR 2022) · 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
- MVCPS-NeuS: Multi-view Constrained Photometric Stereo for Neural Surface Reconstruction (CVPR 2024 · similarity 0.89)
- CNN-PS: CNN-based Photometric Stereo for General Non-Convex Surfaces (ECCV 2018 · similarity 0.89)
- Neural Multi-View Self-Calibrated Photometric Stereo without Photometric Stereo Cues (ICCV 2025 · similarity 0.89)
- PS-NeRF: Neural Inverse Rendering for Multi-View Photometric Stereo (ECCV 2022 · similarity 0.89)
- Learning Efficient Photometric Feature Transform for Multi-View Stereo (ICCV 2021 · similarity 0.89)
- Sparse Views Near Light: A Practical Paradigm for Uncalibrated Point-light Photometric Stereo (CVPR 2024 · similarity 0.87)
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.
