Research map / 3D vision / Depth and stereo
Scalable, Detailed and Mask-Free Universal Photometric Stereo
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Train on PS-Mix rather than PS-Wild (ablation) | UniPS-like local baseline · PS-Wild-Test directional lighting | MAE: 17.0 -> 8.4 | Helped |
| Add non-local interactions among aggregated pixel features (ablation) | SDM-UniPS · PS-Wild-Test directional lighting | MAE: 8.4 -> 5.7 | Helped |
| Increase non-local pixel samples from 2048 to 20480 (ablation) | SDM-UniPS · PS-Wild-Test directional lighting | MAE: 5.7 -> 5.7 | No clear effect |
| Add the scale-invariant spatial-light feature encoder (ablation) | SDM-UniPS with non-local interaction · PS-Wild-Test directional lighting | MAE: 5.7 -> 4.8 | Helped |
| Use fewer input images (ablation) | SDM-UniPS · DiLiGenT | mean angular error (degrees), Ave.: 5.8 -> 6.5 (K=8); 5.8 -> 11.4 (K=2) | Hurt |
| Use SDM-UniPS rather than UniPS | Universal photometric stereo · DiLiGenT | mean angular error (degrees), Ave.: 14.7 -> 5.8 | Helped |
Closest papers by meaning
- Universal Photometric Stereo Network Using Global Lighting Contexts (CVPR 2022 · similarity 0.93)
- Light of Normals: Unified Feature Representation for Universal Photometric Stereo (ICLR 2026 · similarity 0.92)
- Sparse Views Near Light: A Practical Paradigm for Uncalibrated Point-light Photometric Stereo (CVPR 2024 · similarity 0.88)
- A Simple yet Universal Framework for Depth Completion (NeurIPS 2024 · similarity 0.86)
- Stereo Anywhere: Robust Zero-Shot Deep Stereo Matching Even Where Either Stereo or Mono Fail (CVPR 2025 · similarity 0.86)
- MVCPS-NeuS: Multi-view Constrained Photometric Stereo for Neural Surface Reconstruction (CVPR 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.