Research map / 3D vision / Depth and stereo
Deep Single-Image Portrait Relighting
Hao Zhou, Sunil Hadap, Kalyan Sunkavalli, David W. Jacobs
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 the illumination prediction loss branch (ablation) | Our relighting model · Relighting validation set (Target) | RMSE: 0.0460 -> 0.0435 | Improved |
| add the illumination prediction loss branch (ablation) | Our relighting model · Relighting validation set (Target) | RMSE-s: 0.0395 -> 0.0351 | Improved |
| add the illumination prediction loss branch (ablation) | Our relighting model · Relighting validation set (Target) | DSSIM: 0.0251 -> 0.0251 | No clear effect |
| add the self-supervision loss (ablation) | Our relighting model · Relighting validation set (Target) | RMSE: 0.0448 -> 0.0435 | Improved |
| add the self-supervision loss (ablation) | Our relighting model · Relighting validation set (Target) | DSSIM: 0.0258 -> 0.0251 | Improved |
| add the self-supervision loss (ablation) | Our relighting model · Relighting validation set (Source reconstruction) | RMSE: 0.0313 -> 0.0170 | Improved |
| add confidence learning (ablation) | Our lighting estimation · Lighting estimation | RMSE-s: 0.8231 -> 0.6633 | Improved |
| use our model instead of SfSNet for relighting | single-image portrait relighting · Relighting validation set (Target) | RMSE: 0.2397 -> 0.0435 | Improved |
| use our model instead of SIRFS for relighting | single-image portrait relighting · Relighting validation set (Target) | DSSIM: 0.0827 -> 0.0251 | Improved |
| use our model instead of lighting transfer baseline given ground-truth reference (oracle) | lighting transfer · Relighting validation set (lighting transfer comparison) | RMSE: 0.0406 -> 0.0435 | Mixed |
| use our model instead of lighting estimation baselines | lighting estimation · Lighting estimation | RMSE-s: 1.3252 -> 0.6633 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- SfSNet — SfSNet: Learning Shape, Reflectance and Illuminance of Faces `in the Wild' (CVPR 2018) · better on 7 of 7 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- SynthLight: Portrait Relighting with Diffusion Model by Learning to Re-render Synthetic Faces (CVPR 2025 · similarity 0.89)
- Text2Relight: Creative Portrait Relighting with Text Guidance (AAAI 2025 · similarity 0.88)
- Lux Post Facto: Learning Portrait Performance Relighting with Conditional Video Diffusion and a Hybrid Dataset (CVPR 2025 · similarity 0.88)
- Holo-Relighting: Controllable Volumetric Portrait Relighting from a Single Image (CVPR 2024 · similarity 0.87)
- Comprehensive Relighting: Generalizable and Consistent Monocular Human Relighting and Harmonization (CVPR 2025 · similarity 0.87)
- Relightful Harmonization: Lighting-aware Portrait Background Replacement (CVPR 2024 · 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.
