Research map / Image generation / Image restoration
DR2: Diffusion-Based Robust Degradation Remover for Blind Face Restoration
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 diffusion-based degradation removal before SPARNetHD enhancement | SPARNetHD · CelebA-Test 16× | FID: 77.02 -> 53.22 | Helped |
| Add diffusion-based degradation removal before VQFR enhancement | VQFR · CelebA-Test 16× | FID: 152.56 -> 47.29 | Helped |
| Add diffusion-based degradation removal before SPARNetHD enhancement | SPARNetHD · CelebA-Test 8× | FID: 59.66 -> 56.29 | No clear effect |
| Add diffusion-based degradation removal before VQFR enhancement | VQFR · CelebA-Test 8× | FID: 66.54 -> 53.82 | Helped |
| Add diffusion-based degradation removal before SPARNetHD enhancement | SPARNetHD · CelebA-Test 4× | FID: 53.20 -> 51.44 | Helped |
| Add diffusion-based degradation removal before VQFR enhancement | VQFR · CelebA-Test 4× | FID: +0.98 improvement | Helped |
| Add DR2 before VQFR enhancement | VQFR · WIDER-Critical | FID: 81.37 -> 60.06 | Helped |
| Add DR2 before VQFR enhancement | VQFR · WIDER-Normal | FID: 60.84 -> 58.78 | Helped |
| Add DR2 before VQFR enhancement | VQFR · CelebChild | FID: 104.39 -> 103.91 | Helped |
| Add DR2 before VQFR enhancement | VQFR · LFW-Test | FID: 51.81 -> 50.98 | Helped |
| Increase downsampling factor N (ablation) | DR2 · CelebA-Test | PSNR | Mixed |
| Increase truncation step τ (ablation) | DR2 · CelebA-Test | PSNR | Mixed |
Closest papers by meaning
- Resfusion: Denoising Diffusion Probabilistic Models for Image Restoration Based on Prior Residual Noise (NeurIPS 2024 · similarity 0.89)
- Denoising as Adaptation: Noise-Space Domain Adaptation for Image Restoration (ICLR 2025 · similarity 0.88)
- $InterLCM$: Low-Quality Images as Intermediate States of Latent Consistency Models for Effective Blind Face Restoration (ICLR 2025 · similarity 0.88)
- Learning Degradation-unaware Representation with Prior-based Latent Transformations for Blind Face Restoration (CVPR 2024 · similarity 0.88)
- Blind Face Restoration under Extreme Conditions: Leveraging 3D-2D Prior Fusion for Superior Structural and Texture Recovery (AAAI 2024 · similarity 0.88)
- FaceMe: Robust Blind Face Restoration with Personal Identification (AAAI 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.