Research map / 3D vision / Depth and stereo
Controllable Light Diffusion for Portraits
David Futschik, Kelvin Ritland, James Vecore, Sean Fanello, Sergio Orts-Escolano, Brian Curless, Daniel Sýkora, Rohit Pandey
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 synthetic external shadow augmentation (ablation) | Light diffusion model · Light stage dataset | MAE: 0.0123 -> 0.0098 | Improved |
| add synthetic external shadow augmentation (ablation) | Light diffusion model · Light stage dataset | MSE: 0.0009 -> 0.0006 | Improved |
| add synthetic external shadow augmentation (ablation) | Light diffusion model · Light stage dataset | SSIM: 0.9563 -> 0.9692 | Improved |
| add synthetic external shadow augmentation (ablation) | Light diffusion model · Light stage dataset | LPIPS: 0.0503 -> 0.0340 | Improved |
| use twice as many filters and omit shadow/specular maps (ablation) | Light diffusion model · Light stage dataset | MAE: 0.0098 -> 0.0094 | Improved |
| use twice as many filters and omit shadow/specular maps (ablation) | Light diffusion model · Light stage dataset | MSE: 0.0006 -> 0.0005 | Improved |
| use twice as many filters and omit shadow/specular maps (ablation) | Light diffusion model · Light stage dataset | SSIM: 0.9692 -> 0.9749 | Improved |
| use twice as many filters and omit shadow/specular maps (ablation) | Light diffusion model · Light stage dataset | LPIPS: 0.0340 -> 0.0304 | Improved |
| use twice as many filters and omit shadow/specular maps (ablation) | Light diffusion model · In-the-wild portraits | shadow-edge quality | Hurt |
Closest papers by meaning
- SynthLight: Portrait Relighting with Diffusion Model by Learning to Re-render Synthetic Faces (CVPR 2025 · similarity 0.91)
- LightIt: Illumination Modeling and Control for Diffusion Models (CVPR 2024 · similarity 0.88)
- Lux Post Facto: Learning Portrait Performance Relighting with Conditional Video Diffusion and a Hybrid Dataset (CVPR 2025 · similarity 0.88)
- Relightful Harmonization: Lighting-aware Portrait Background Replacement (CVPR 2024 · similarity 0.88)
- Neural Gaffer: Relighting Any Object via Diffusion (NeurIPS 2024 · similarity 0.87)
- PHR-DIFF: Portrait Highlights Removal via Patch-aware Diffusion Model (AAAI 2025 · similarity 0.87)
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.
