Research map / 3D vision / Depth and stereo
Photometric Stereo via Discrete Hypothesis-and-Test Search
Kenji Enomoto, Michael Waechter, Kiriakos N. Kutulakos, Yasuyuki Matsushita
Check 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 |
|---|---|---|---|
| Use exhaustive discrete normal search with BRDF tensor precomputation | Photometric stereo on MERL sphere · MERL sphere (100 materials) | mean angular error (average/variance across materials): 2.23/3.24 vs 3.15/9.19 | Improved |
| Use exhaustive discrete normal search with BRDF tensor precomputation | Photometric stereo · DiLiGenT average | mean angular error: 9.53 vs 7.20 | Hurt |
| Reduce the number of normal candidates from 20001 to 10001 (ablation) | Proposed photometric stereo method · MERL sphere (5-fold cross-validation) | mean angular error: 2.25 -> 2.31 | Hurt |
| Reduce the number of normal candidates from 20001 to 1501 (ablation) | Proposed photometric stereo method · MERL sphere (5-fold cross-validation) | mean angular error: 2.25 -> 2.86 | Hurt |
| Reduce the number of normal candidates from 20001 to 251 (ablation) | Proposed photometric stereo method · MERL sphere (5-fold cross-validation) | mean angular error: 2.25 -> 4.62 | Hurt |
| Reduce the number of lights from 100 to 20 (ablation) | Proposed photometric stereo method · MERL sphere (5-fold cross-validation) | mean angular error: 2.22 -> 2.69 | Hurt |
| Set reduced dimension M′ to 5 instead of 3 (ablation) | Proposed photometric stereo method · Noiseless MERL sphere, 80 lights | mean angular error: 2.18 -> 1.76 | Improved |
| Set reduced dimension M′ to 5 instead of 3 (ablation) | Proposed photometric stereo method · Noisy MERL sphere, 80 lights, noise 5/30 | mean angular error: 3.99 -> 4.12 | Hurt |
| Use discretized light directions closest to the known lights (ablation) | Proposed photometric stereo method · MERL sphere, 100 lights | mean angular error: 2.15 -> 2.16 | No clear effect |
Closest papers by meaning
- Discrete Search Photometric Stereo for Fast and Accurate Shape Estimation (TPAMI 2022 · similarity 0.96)
- Shape and Spatially-Varying Reflectance Estimation from Virtual Exemplars (TPAMI 2016 · similarity 0.92)
- A Microfacet-Based Reflectance Model for Photometric Stereo With Highly Specular Surfaces (ICCV 2017 · similarity 0.89)
- Learning to Minify Photometric Stereo (CVPR 2019 · similarity 0.89)
- Neural Inverse Rendering for General Reflectance Photometric Stereo (ICML 2018 · similarity 0.89)
- Uncalibrated Photometric Stereo by Stepwise Optimization Using Principal Components of Isotropic BRDFs (CVPR 2016 · 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.
