Research map / 3D vision / Depth and stereo
Materials for Masses: SVBRDF Acquisition with a Single Mobile Phone Image
Zhengqin Li, Kalyan Sunkavalli, Manmohan Chandraker
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 a material classifier whose outputs weight per-material BRDF predictions (ablation) | encoder-decoder · SVBRDF test | albedo L2 error: 7.78 -> 7.58 | Improved |
| add a material classifier (ablation) | encoder-decoder · SVBRDF test | normal L2 error: 1.55 -> 1.52 | Improved |
| add a material classifier (ablation) | encoder-decoder · SVBRDF test | roughness L2 error: 8.75 -> 8.55 | Improved |
| add DCRF refinement on top of the classifier (ablation) | cls-pt · SVBRDF test | albedo L2 error: 7.58 -> 7.42 | Improved |
| add DCRF refinement on top of the classifier (ablation) | cls-pt · SVBRDF test | normal L2 error: 1.52 -> 1.50 | Improved |
| add DCRF refinement on top of the classifier (ablation) | cls-pt · SVBRDF test | roughness L2 error: 8.55 -> 8.53 | Improved |
| train the classifier jointly with the BRDF network instead of as a pure classifier (ablation) | material classifier · material classification | classification error (%): 54.96 -> 73.65 | Improved |
| add a classifier but do not use its output to fuse BRDF predictions (ablation) | encoder-decoder · SVBRDF test | BRDF estimation error | No clear effect |
| add a collocated point (flash) light to environment lighting at capture (ablation) | cls network · SVBRDF test | SVBRDF error | Improved |
| reduce relative flash intensity against environment light (ablation) | full network · SVBRDF test | SVBRDF error | Hurt |
| use the proposed single-image network instead of the two-shot optimization method | SVBRDF reconstruction · stationary texture 192x192 | reconstruction time: 6 hours -> 2.4 seconds | Improved |
Closest papers by meaning
- Reflectance Capture Using Univariate Sampling of BRDFs (ICCV 2017 · similarity 0.89)
- Material Palette: Extraction of Materials from a Single Image (CVPR 2024 · similarity 0.89)
- Single-Shot Neural Relighting and SVBRDF Estimation (ECCV 2020 · similarity 0.89)
- On Joint Estimation of Pose, Geometry and svBRDF From a Handheld Scanner (CVPR 2020 · similarity 0.89)
- Intrinsic Image Diffusion for Indoor Single-view Material Estimation (CVPR 2024 · similarity 0.84)
- Real-Time Neural BRDF with Spherically Distributed Primitives (CVPR 2024 · similarity 0.84)
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.
