Research map / Image generation / Image restoration
RestoreFormer: High-Quality Blind Face Restoration From Undegraded Key-Value Pairs
Zhouxia Wang, Jiawei Zhang, Runjian Chen, Wenping Wang, Ping Luo
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 self-attention on degraded-image features instead of no attention (ablation) | RestoreFormer · CelebA-Test | FID: 50.68 -> 47.39 | Improved |
| use self-attention on degraded-image features instead of no attention (ablation) | RestoreFormer · CelebA-Test | IDD: 0.6401 -> 0.6284 | Improved |
| replace MHCA with local SFT fusion (ablation) | RestoreFormer · CelebA-Test | FID: 41.45 -> 41.47 | Hurt |
| replace MHCA with local SFT fusion (ablation) | RestoreFormer · CelebA-Test | IDD: 0.5650 -> 0.6702 | Hurt |
| use only degraded-image features with MHSA (ablation) | RestoreFormer · CelebA-Test | IDD: 0.5650 -> 0.6284 | Improved |
| use only dictionary priors with MHSA (ablation) | RestoreFormer · CelebA-Test | FID: 41.45 -> 45.83 | Improved |
| use only dictionary priors with MHSA (ablation) | RestoreFormer · CelebA-Test | IDD: 0.5650 -> 0.7662 | Hurt |
| fuse degraded-image and prior information with MHCA rather than using either source alone (ablation) | RestoreFormer · CelebA-Test | FID: 45.83 -> 41.45 | Improved |
| fuse degraded-image and prior information with MHCA rather than using either source alone (ablation) | RestoreFormer · CelebA-Test | IDD: 0.6284 -> 0.5650 | Improved |
| use degraded features rather than priors in the MHCA skip connection (ablation) | RestoreFormer · CelebA-Test | FID: 41.45 -> 42.00 | Hurt |
| use degraded features rather than priors in the MHCA skip connection (ablation) | RestoreFormer · CelebA-Test | IDD: 0.5650 -> 0.5938 | Improved |
| replace the reconstruction-oriented dictionary with a recognition-oriented VGG dictionary (ablation) | RestoreFormer · CelebA-Test | FID: 41.45 -> 61.43 | Hurt |
Later papers that compared with this method
7 later accepted papers report RestoreFormer++ as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.
- Posterior-Mean Rectified Flow: Towards Minimum MSE Photo-Realistic Image Restoration (ICLR 2025 · better on 0 of 1 shared result)
- DynFaceRestore: Balancing Fidelity and Quality in Diffusion-Guided Blind Face Restoration with Dynamic Blur-Level Mapping and Guidance (ICCV 2025 · better on 0 of 1 shared result)
- Unlocking the Potential of Diffusion Priors in Blind Face Restoration (ICCV 2025 · better on 0 of 1 shared result)
- Learning Degradation-unaware Representation with Prior-based Latent Transformations for Blind Face Restoration (CVPR 2024 · better on 2 of 2 shared results)
- WaveFace: Authentic Face Restoration with Efficient Frequency Recovery (CVPR 2024 · better on 1 of 2 shared results)
- FlowIE: Efficient Image Enhancement via Rectified Flow (CVPR 2024 · better on 0 of 1 shared result)
- CosAE: Learnable Fourier Series for Image Restoration (NeurIPS 2024 · better on 1 of 1 shared result)
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.
- PULSE — PULSE: Self-Supervised Photo Upsampling via Latent Space Exploration of Generative Models (CVPR 2020) · 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
- RestoreFormer++: Towards Real-World Blind Face Restoration From Undegraded Key-Value Pairs (TPAMI 2023 · similarity 0.96)
- Towards Robust Blind Face Restoration with Codebook Lookup Transformer (NeurIPS 2022 · similarity 0.90)
- FaceMe: Robust Blind Face Restoration with Personal Identification (AAAI 2025 · similarity 0.88)
- Blind Face Restoration under Extreme Conditions: Leveraging 3D-2D Prior Fusion for Superior Structural and Texture Recovery (AAAI 2024 · similarity 0.87)
- RetouchFormer: Semi-supervised High-Quality Face Retouching Transformer with Prior-Based Selective Self-Attention (AAAI 2024 · similarity 0.86)
- Sharing Key Semantics in Transformer Makes Efficient Image Restoration (NeurIPS 2024 · 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.
