Research map / Image generation / Super-resolution
Interpreting Super-Resolution Networks With Local Attribution Maps
Check 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 information from a wider range of input pixels | SR networks · 48 SR networks evaluated on the challenging image set | PSNR: both larger than 0.8 | Helped |
| Increase FSRCNN depth and width | FSRCNN · SR performance and DI analysis | PSNR | Helped |
| Increase EDSR depth by adding convolution layers | EDSR · SR performance and DI analysis | PSNR | Helped |
| Increase the number of feature maps per EDSR layer | EDSR · SR performance and DI analysis | PSNR | No clear effect |
| Add global operations without a sufficiently large effective receptive field | IMDN, RFDN, and RNAN · SR test set (150 challenging images) | DI | No clear effect |
| Use attention or non-local operations to access distant input pixels | Attention and non-local SR networks · SR image patches | DI | Helped |
| Rely on a wider area of pixels without using that information effectively | SAN and DRLN · SR image patches | PSNR | Hurt |
| Use a blurred image as the attribution baseline and progressive blurring as the path | LAM for SR networks · SR attribution analysis | Attribution validity | Helped |
Closest papers by meaning
- From Coarse to Fine: Hierarchical Pixel Integration for Lightweight Image Super-resolution (AAAI 2023 · similarity 0.90)
- Accurate Image Super-Resolution Using Very Deep Convolutional Networks (CVPR 2016 · similarity 0.90)
- ADD: Attribution-Driven Data Augmentation Framework for Boosting Image Super-Resolution (CVPR 2025 · similarity 0.89)
- Perception-Oriented Single Image Super-Resolution Using Optimal Objective Estimation (CVPR 2023 · similarity 0.88)
- Image Processing GNN: Breaking Rigidity in Super-Resolution (CVPR 2024 · similarity 0.87)
- SeD: Semantic-Aware Discriminator for Image Super-Resolution (CVPR 2024 · 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.