Research map / Image generation / Editing and style transfer
PASDiff: Physics-Aware Semantic Guidance for Joint Real-world Low-Light Face Enhancement and Restoration
Yilin Ni, Wenjie Li, Zhengxue Wang, Juncheng Li, Guangwei Gao, Jian Yang
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 exposure loss L_exp (ablation) | PASDiff · FFHQ (synthetic) | visual brightness | Improved |
| add Retinex reflectance loss L_ref (ablation) | PASDiff · FFHQ (synthetic) | color recovery | Improved |
| add structural guidance loss L_stru (ablation) | PASDiff · FFHQ (synthetic) | LPIPS and LMD | Improved |
| add structural guidance loss L_stru (ablation) | PASDiff · FFHQ (synthetic) | PSNR | Hurt |
| replace MSE loss to prior output with AdaIN-based SASI (ablation) | PASDiff · FFHQ (synthetic) | color and global tone | Improved |
| joint training-free PASDiff instead of best cascade LightenDiffusion -> DiffBIR | PASDiff · FFHQ (synthetic) | PSNR: 18.20 -> 22.59 | Improved |
| joint PASDiff instead of best cascade | PASDiff · WildDark-Face (real) | FID: 121.47 vs 127.04 | Hurt |
| PASDiff versus second-best method in identity matching | PASDiff · FFHQ (face recognition) | face recognition accuracy: over 8% higher | Improved |
| order of cascade: enhance then restore vs restore then enhance | LightenDiffusion + DiffBIR · WildDark-Face (real) | MUSIQ: 49.4678 vs 49.1503 | 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.
- LEDNet — LEDNet: Joint Low-Light Enhancement and Deblurring in the Dark (ECCV 2022) · better on 5 of 5 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
- Low-Light Face Super-resolution via Illumination, Structure, and Texture Associated Representation (AAAI 2024 · similarity 0.87)
- Self-Supervised Selective-Guided Diffusion Model for Old-Photo Face Restoration (NeurIPS 2025 · similarity 0.87)
- Blind Face Restoration under Extreme Conditions: Leveraging 3D-2D Prior Fusion for Superior Structural and Texture Recovery (AAAI 2024 · similarity 0.86)
- $InterLCM$: Low-Quality Images as Intermediate States of Latent Consistency Models for Effective Blind Face Restoration (ICLR 2025 · similarity 0.86)
- AGLLDiff: Guiding Diffusion Models Towards Unsupervised Training-free Real-world Low-light Image Enhancement (AAAI 2025 · similarity 0.85)
- Luminance-Aware Statistical Quantization: Unsupervised Hierarchical Learning for Illumination Enhancement (NeurIPS 2025 · similarity 0.85)
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.
