Research map / 3D vision / Depth and stereo
Learning Physics-Guided Face Relighting Under Directional Light
Thomas Nestmeyer, Jean-Francois Lalonde, Iain Matthews, Andreas Lehrmann
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 physics-guided intrinsic decomposition and non-diffuse residual refinement instead of pix2pix | Face relighting model · Face relighting test, with source illumination | DSSIM: .1832 -> .1618 | Improved |
| Use physics-guided intrinsic decomposition and non-diffuse residual refinement instead of pix2pix | Face relighting model · Face relighting test, without source illumination | DSSIM: .2076 -> .1763 | Improved |
| Remove explicit access to source illumination (ablation) | Ours · Face relighting test | DSSIM: .1618 -> .1763 | Hurt |
| Train with DSSIM rather than L1 loss (ablation) | Ours · Validation face relighting | DSSIM: .1484 -> .1270 | Improved |
| Train with DSSIM rather than L2 loss (ablation) | Ours · Validation face relighting | DSSIM: .1547 -> .1270 | Improved |
| Train with LPIPS rather than DSSIM loss (ablation) | Ours · Validation face relighting | LPIPS: .2008 -> .1701 | Hurt |
| Add image and lighting intensity scaling augmentation (ablation) | Training data · Face relighting | unspecified | No clear effect |
Closest papers by meaning
- SwitchLight: Co-design of Physics-driven Architecture and Pre-training Framework for Human Portrait Relighting (CVPR 2024 · similarity 0.91)
- SynthLight: Portrait Relighting with Diffusion Model by Learning to Re-render Synthetic Faces (CVPR 2025 · similarity 0.91)
- Comprehensive Relighting: Generalizable and Consistent Monocular Human Relighting and Harmonization (CVPR 2025 · similarity 0.89)
- Lux Post Facto: Learning Portrait Performance Relighting with Conditional Video Diffusion and a Hybrid Dataset (CVPR 2025 · similarity 0.89)
- UniLumos: Fast and Unified Image and Video Relighting with Physics-Plausible Feedback (NeurIPS 2025 · similarity 0.88)
- UniRelight: Learning Joint Decomposition and Synthesis for Video Relighting (NeurIPS 2025 · 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.
