Research map / Image generation / Image restoration
After the Party: Navigating the Mapping From Color to Ambient Lighting
Florin-Alexandru Vasluianu, Tim Seizinger, Zongwei Wu, Radu Timofte
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 HSV guidance with CDFFA fusion to feature refinement (ablation) | RLN2 · CL3AN test split | PSNR: 19.892 -> 20.523 | Improved |
| use HSV rather than RGB as guidance representation (concat fusion) (ablation) | RLN2 · CL3AN test split | PSNR: 20.018 -> 20.216 | Improved |
| use HSV rather than L*a*b* guidance (concat fusion) (ablation) | RLN2 · CL3AN test split | LPIPS: 0.226 -> 0.220 | Improved |
| replace concatenation fusion with CDFFA attention fusion (HSV guidance) (ablation) | RLN2 · CL3AN test split | PSNR: 20.216 -> 20.523 | Improved |
| use CDFFA instead of concatenation on compute (ablation) | RLN2 · CL3AN test split | MACs (G): 23.68 -> 22.72 | Improved |
| add frequency-domain information to RLN2 (ablation) | RLN2-L · CL3AN test split | PSNR: 20.383 -> 20.523 | Improved |
| use large ConvNeXt-XL context extractor instead of ConvNeXt-T (ablation) | RLN2 · CL3AN test split | PSNR: 19.992 -> 20.383 | Improved |
| use state space models (MAMBAIR) for ambient lighting normalization | baseline models · CL3AN test split | PSNR: 18.970 | Hurt |
| use RLN2-Lf instead of IFBlend | IFBlend · AMBIENT6K | PSNR: 21.443 -> 21.712 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.
- Restormer — Restormer: Efficient Transformer for High-Resolution Image Restoration (CVPR 2022) · better on 2 of 2 shared results
- Retinexformer — Retinexformer: One-stage Retinex-based Transformer for Low-light Image Enhancement (ICCV 2023) · better on 1 of 1 shared result
- HINet — HiNet: Deep Image Hiding by Invertible Network (ICCV 2021) · better on 1 of 1 shared result
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
- Eclipse: Disambiguating Illumination and Materials using Unintended Shadows (CVPR 2024 · similarity 0.87)
- UniRelight: Learning Joint Decomposition and Synthesis for Video Relighting (NeurIPS 2025 · similarity 0.87)
- Recasting Regional Lighting for Shadow Removal (AAAI 2024 · similarity 0.87)
- Factored-NeuS: Reconstructing Surfaces, Illumination, and Materials of Possibly Glossy Objects (CVPR 2025 · similarity 0.87)
- MetaGS: A Meta-Learned Gaussian-Phong Model for Out-of-Distribution 3D Scene Relighting (NeurIPS 2025 · similarity 0.87)
- Bright-NeRF: Brightening Neural Radiance Field with Color Restoration from Low-Light RAW Images (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.
