Research map / Image generation / Image restoration
Face2Scene: Using Facial Degradation as an Oracle for Diffusion-Based Scene Restoration
Amirhossein Kazerouni, Maitreya Suin, Tristan Aumentado-Armstrong, Sina Honari, Amanpreet Walia, Iqbal Mohomed, Konstantinos G. Derpanis, Babak Taati, Alex Levinshtein
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 face-derived degradation estimation (FaDeX + MapNet) (ablation) | Face2Scene · InScene Synthetic Validation | DISTS / FID: DISTS 0.1293 -> 0.1007; FID 17.81 -> 15.26 | Improved |
| add face-derived degradation estimation (ablation) | Face2Scene · InScene Synthetic Validation | PSNR / SSIM: PSNR 22.7160 -> 22.9040; SSIM 0.5476 -> 0.5574 | Mixed |
| Face2Scene instead of S3Diff (face-derived vs LQ-only degradation) | one-step SD-Turbo restorer · InScene Real Validation | LPIPS: 0.5149 -> 0.2502 | Improved |
| Face2Scene instead of S3Diff on pixel fidelity | restorer · InScene Synthetic Validation | PSNR: 23.5955 -> 22.9040 | Hurt |
| Face2Scene versus S3Diff with GT face composited in (non-face quality) | restorer · InScene Synthetic Validation | DISTS: 0.1000 -> 0.0908 | Improved |
| Face2Scene versus S3Diff on real validation FID | restorer · InScene Real Validation | FID: 38.64 -> 42.21 | Hurt |
| add Stage I Ref-FR and degradation extraction cost | Face2Scene · 1024x1024 image runtime | seconds: 1.4 (S3Diff) -> 3.3 | Hurt |
Closest papers by meaning
- Self-Supervised Selective-Guided Diffusion Model for Old-Photo Face Restoration (NeurIPS 2025 · similarity 0.89)
- Overcoming False Illusions in Real-World Face Restoration with Multi-Modal Guided Diffusion Model (ICLR 2025 · similarity 0.89)
- OSDFace: One-Step Diffusion Model for Face Restoration (CVPR 2025 · similarity 0.88)
- ReF-LDM: A Latent Diffusion Model for Reference-based Face Image Restoration (NeurIPS 2024 · similarity 0.88)
- FaceMe: Robust Blind Face Restoration with Personal Identification (AAAI 2025 · similarity 0.88)
- Learning Degradation-unaware Representation with Prior-based Latent Transformations for Blind Face Restoration (CVPR 2024 · 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.
