Research map / 3D vision / Depth and stereo
Don’t Mask Out the Background! Natural-Light Photometric Stereo via Illumination Reconstruction
Taiga Hashida, Hiroaki Santo, Fumio Okura
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 |
|---|---|---|---|
| reconstruct illumination from background with 3DGS and do inverse rendering | NaPS · Synthetic (avg) | normal MAngE: 4.92 (SDM-UniPS) -> 3.35 | Improved |
| same | NaPS · Synthetic (avg) | albedo PSNR: 28.4 (LINO-UniPS) -> 32.5 | Improved |
| same | NaPS · Synthetic (avg) | roughness MAbsE: 0.0659 (LINO-UniPS) -> 0.0610 | Mixed |
| same | NaPS · Synthetic BEAR/BUDDHA | albedo PSNR / roughness: BEAR albedo 34.4 -> 30.2 | Mixed |
| increase number of views from 40 to 150 (ablation) | Ours · Synthetic | normal MAngE: 5.10 -> 3.35 | Improved |
| increase number of views from 40 to 150 (ablation) | SDM-UniPS · Synthetic | normal MAngE: 4.84 -> 4.92 | No clear effect |
| add emitter-guided light sampling (ablation) | Ours · Synthetic ablation | normal MAngE: 3.78 -> 3.35 | Improved |
| add emitter-guided light sampling (ablation) | Ours · Synthetic ablation | roughness MAbsE: 0.0731 -> 0.0610 | Improved |
| add two-stage 3D Gaussian emitter optimization (ablation) | Ours · Synthetic ablation | albedo PSNR: 25.9 -> 32.5 | Improved |
| add two-stage 3D Gaussian emitter optimization (ablation) | Ours · Synthetic ablation | normal MAngE: 3.59 -> 3.35 | Improved |
Closest papers by meaning
- Self-Calibrating Photometric Stereo by Neural Inverse Rendering (ECCV 2022 · similarity 0.91)
- Sparse Views Near Light: A Practical Paradigm for Uncalibrated Point-light Photometric Stereo (CVPR 2024 · similarity 0.90)
- DeepPS2: Revisiting Photometric Stereo Using Two Differently Illuminated Images (ECCV 2022 · similarity 0.90)
- Neural Multi-View Self-Calibrated Photometric Stereo without Photometric Stereo Cues (ICCV 2025 · similarity 0.89)
- Uncalibrated Neural Inverse Rendering for Photometric Stereo of General Surfaces (CVPR 2021 · similarity 0.89)
- Spin-UP: Spin Light for Natural Light Uncalibrated Photometric Stereo (CVPR 2024 · similarity 0.88)
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.
