Research map / Image generation / Super-resolution
VarSR: Variational Super-Resolution Network for Very Low Resolution Images
Sangeek Hyun, Jae-Pil Heo
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 |
|---|---|---|---|
| raise lambda_KL from 1e-2 to 2e-2 (no adversarial loss) (ablation) | VarSR-Net · CelebA | PSNR: 22.46 -> 22.74 | Improved |
| raise lambda_KL from 1e-2 to 2e-2 (with adversarial loss) (ablation) | VarSR-Net · CelebA | PSNR: 21.87 -> 22.14 | Improved |
| raise lambda_KL from 1e-2 to 2e-2 (ablation) | VarSR-Net · CelebA | diversity (avg LPIPS): 0.0353 -> 0.0238 | Hurt |
| raise lambda_KL from 1e-2 to 2e-2 (no adversarial loss) (ablation) | VarSR-Net · CelebA | LPIPS: 0.0885 -> 0.0927 | Hurt |
| add adversarial loss (ablation) | VarSR-Net (lambda_KL=2e-2) · CelebA | LPIPS: 0.0927 -> 0.0538 | Improved |
| add adversarial loss (ablation) | VarSR-Net (lambda_KL=2e-2) · CelebA | PSNR: 22.74 -> 22.14 | Hurt |
| use VarSR-Net instead of deterministic model | autoencoder SR · MNIST | PSNR (best of 5): 20.64 -> 22.00 | Improved |
| use VarSR-Net instead of deterministic model | autoencoder SR · LP | PSNR (best of 5): 21.16 -> 22.00 | Improved |
| use VarSR-Net instead of deterministic model | autoencoder SR · MNIST | classification accuracy, Best: 84.74 -> 92.79 | Improved |
| use VarSR-Net instead of deterministic model | autoencoder SR · LP | classification accuracy, Best: 93.14 -> 96.33 | Improved |
| use VarSR-Net instead of deterministic model | autoencoder SR · MNIST | classification accuracy, Mean: 84.74 -> 85.58 | No clear effect |
| use VarSR-Net instead of SRGAN (with adversarial loss) | SRGAN · CelebA | PSNR: 22.28 -> 21.87 | Hurt |
Closest papers by meaning
- Explorable Super Resolution (CVPR 2020 · similarity 0.90)
- Natural and Realistic Single Image Super-Resolution With Explicit Natural Manifold Discrimination (CVPR 2019 · similarity 0.89)
- Visual Autoregressive Modeling for Image Super-Resolution (ICML 2025 · similarity 0.88)
- Perception-Oriented Single Image Super-Resolution Using Optimal Objective Estimation (CVPR 2023 · similarity 0.88)
- Noise-Free Optimization in Early Training Steps for Image Super-resolution (AAAI 2024 · similarity 0.88)
- TSD-SR: One-Step Diffusion with Target Score Distillation for Real-World Image Super-Resolution (CVPR 2025 · similarity 0.87)
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.
