Research map / 3D vision / Face and shape
Inverse Rendering of Translucent Objects Using Physical and Neural Renderers
Chenhao Li, Trung Thanh Ngo, Hajime Nagahara
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 physical surface-reflectance renderer and neural SSS renderer instead of autoencoder reconstruction (ablation) | Inverse rendering model · Synthetic test scenes (17140) | spherical harmonics MAE: .1083(.6571) -> .1064(.8984) | Improved |
| Add a physical surface-reflectance renderer and neural SSS renderer instead of autoencoder reconstruction (ablation) | Inverse rendering model · Synthetic test scenes (17140) | roughness MAE: .0811(.5303) -> .0811(.4903) | No clear effect |
| Add augmented supervision with images rendered using edited SSS parameters (ablation) | 2R model · Synthetic test scenes (17140) | extinction coefficient MAE: .1675(1.072) -> .1619(.9635) | Improved |
| Add augmented supervision with images rendered using edited SSS parameters (ablation) | 2R model · Synthetic test scenes (17140) | spherical harmonics MAE: .1064(.8984) -> .1105(8.926) | Hurt |
| Add flash and no-flash two-shot input (ablation) | 2R-AUG model · Synthetic test scenes (17140) | phase function parameter MAE: .1703(.8856) -> .1655(.3932) | Improved |
| Add flash and no-flash two-shot input (ablation) | 2R-AUG model · Synthetic test scenes (17140) | depth MAE: .0699(1.271) -> .0646(.3283) | Improved |
| Use joint shape, reflectance, illumination, and SSS estimation instead of pure SSS estimation | Translucent inverse rendering · Synthetic test scenes (17140) | extinction coefficient MAE: .1828 -> .1590(.2286) | Improved |
Closest papers by meaning
- Inverse Rendering of Glossy Objects via the Neural Plenoptic Function and Radiance Fields (CVPR 2024 · similarity 0.90)
- RobIR: Robust Inverse Rendering for High-Illumination Scenes (NeurIPS 2024 · similarity 0.89)
- Factored-NeuS: Reconstructing Surfaces, Illumination, and Materials of Possibly Glossy Objects (CVPR 2025 · similarity 0.89)
- Diffusion Renderer: Neural Inverse and Forward Rendering with Video Diffusion Models (CVPR 2025 · similarity 0.89)
- Eclipse: Disambiguating Illumination and Materials using Unintended Shadows (CVPR 2024 · similarity 0.88)
- Differentiable Neural Surface Refinement for Modeling Transparent Objects (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.
