Research map / Image generation / Image restoration
Unlocking the Potential of Diffusion Priors in Blind Face Restoration
Yunqi Miao, Zhiyu Qu, Mingqi Gao, Changrui Chen, Jifei Song, Jungong Han, Jiankang Deng
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 off-shelf degraded images from degradation mode to training (ablation) | Ours · LFW | FID: 43.10 -> 42.98 | Improved |
| add off-shelf degraded images from degradation mode to training (ablation) | Ours · WebPhoto | FID: 83.34 -> 81.37 | Improved |
| add off-shelf degraded images from degradation mode to training (ablation) | Ours · WIDER | FID: 33.09 -> 31.81 | Improved |
| set enhance weight to 1 instead of 0 (ablation) | Ours restoration mode · CelebA-Test | identity fidelity | Improved |
| set enhance weight to 1 instead of 2 (ablation) | Ours restoration mode · CelebA-Test | detail quality | Improved |
| add Association stage when training BFR-oriented face embeddings (ablation) | Ours restoration mode · CelebA-Test | restoration quality | Improved |
| use BFR-oriented face embeddings instead of ArcFace embeddings (ablation) | Ours restoration mode · CelebA-Test | facial detail | Improved |
| replace Real-ESRGAN degradation with learned degradation mode (ablation) | Ours · real-world feature distribution | degradation diversity | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 2 on most shared results.
- WaveFace — WaveFace: Authentic Face Restoration with Efficient Frequency Recovery (CVPR 2024) · better on 1 of 1 shared result
- RestoreFormer++ — RestoreFormer: High-Quality Blind Face Restoration From Undegraded Key-Value Pairs (CVPR 2022) · better on 0 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
- $InterLCM$: Low-Quality Images as Intermediate States of Latent Consistency Models for Effective Blind Face Restoration (ICLR 2025 · similarity 0.91)
- ReF-LDM: A Latent Diffusion Model for Reference-based Face Image Restoration (NeurIPS 2024 · similarity 0.90)
- OSDFace: One-Step Diffusion Model for Face Restoration (CVPR 2025 · similarity 0.89)
- Learning Degradation-unaware Representation with Prior-based Latent Transformations for Blind Face Restoration (CVPR 2024 · similarity 0.88)
- WaveFace: Authentic Face Restoration with Efficient Frequency Recovery (CVPR 2024 · similarity 0.88)
- Overcoming False Illusions in Real-World Face Restoration with Multi-Modal Guided Diffusion Model (ICLR 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.
