Research map / Image generation / Super-resolution
Facial Attribute Capsules for Noise Face Super Resolution
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 |
|---|---|---|---|
| use FACN instead of AEUN | Face super-resolution · CelebA BicN | PSRN: 24.24 -> 24.61 | Helped |
| use FACN instead of AEUN | Face super-resolution · CelebA BBicN | SSIM: 0.6514 -> 0.6714 | Helped |
| use FACN instead of Wavelet-SRNet | Face super-resolution identity retention · BBicN face verification | Performance: 0.8820 -> 0.8922 | Helped |
| add probabilistic capsules to semantic-capsule FACN (ablation) | FACN · CelebA BicN | visual SR quality | Helped |
| add semantic capsules to probabilistic-capsule FACN (ablation) | FACN · CelebA BicN | visual SR quality | Helped |
| replace classical capsules with facial attribute capsules (ablation) | FACN · CelebA BicN | visual SR quality | Helped |
| increase capsule number from k=18 to k=32 (ablation) | FACN · CelebA BicN | PSNR: 24.25 -> 24.49 | Helped |
| increase capsule number from k=32 to k=64 (ablation) | FACN · CelebA BicN | PSNR: 24.49 -> 24.61 | Helped |
| increase capsule number from k=64 to k=128 (ablation) | FACN · CelebA BicN | PSNR: 24.61 -> 24.68 | Helped |
| increase capsule dimension from d=2 to d=4 (ablation) | FACN · CelebA BicN | PSNR: 24.45 -> 24.61 | Helped |
| increase capsule dimension from d=4 to d=8 (ablation) | FACN · CelebA BicN | PSNR: 24.61 -> 24.69 | Helped |
| increase capsule dimension from d=8 to d=16 (ablation) | FACN · CelebA BicN | PSNR: 24.69 -> 24.75 | Helped |
Closest papers by meaning
- Super-Resolving Very Low-Resolution Face Images With Supplementary Attributes (CVPR 2018 · similarity 0.88)
- Residual Attribute Attention Network for Face Image Super-Resolution (AAAI 2019 · similarity 0.88)
- FSRNet: End-to-End Learning Face Super-Resolution With Facial Priors (CVPR 2018 · similarity 0.87)
- Low-Light Face Super-resolution via Illumination, Structure, and Texture Associated Representation (AAAI 2024 · similarity 0.84)
- Vec2Face: Scaling Face Dataset Generation with Loosely Constrained Vectors (ICLR 2025 · similarity 0.83)
- FaceBench: A Multi-View Multi-Level Facial Attribute VQA Dataset for Benchmarking Face Perception MLLMs (CVPR 2025 · similarity 0.82)
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.