Research map / Image generation / Super-resolution
ShuffleMixer: An Efficient ConvNet for Image 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 channel splitting and shuffling for projection (ablation) | ShuffleMixer baseline · DIV2K validation ×4 | PSNR: 29.96 -> 29.83 | Hurt |
| Repeat the CSS-based projection layer (ablation) | ShuffleMixer baseline · DIV2K validation ×4 | PSNR: 29.83 -> 29.99 | Helped |
| Add a 3 × 3 convolution after every two shuffle mixer layers (ablation) | CDC · DIV2K validation ×4 | PSNR: 29.99 -> 30.12 | Helped |
| Fuse input and output by summation, then apply a 3 × 3 convolution (ablation) | CDC · DIV2K validation ×4 | PSNR: 29.99 -> 30.16 | Helped |
| Concatenate input and output, then apply a 3 × 3 convolution (ablation) | CDC · DIV2K validation ×4 | PSNR: 29.99 -> 30.20 | Helped |
| Replace S-Conv with residual blocks (ablation) | CDC · DIV2K validation ×4 | PSNR: 30.16 -> 30.24 | Helped |
| Replace S-Conv with Fused-MBConv (ablation) | CDC · DIV2K validation ×4 | PSNR: 30.16 -> 30.21 | Helped |
| Increase depth-wise kernel size from 3 × 3 to 7 × 7 (ablation) | ShuffleMixer · DIV2K validation ×4 | PSNR: 30.21 -> 30.28 | Helped |
| Increase depth-wise kernel size from 7 × 7 to 9 × 9 (ablation) | ShuffleMixer · DIV2K validation ×4 | PSNR: 30.28 -> 30.29 | No clear effect |
| Increase depth-wise kernel size from 7 × 7 to 11 × 11 (ablation) | ShuffleMixer · DIV2K validation ×4 | PSNR: 30.28 -> 30.28 | No clear effect |
| Increase depth-wise kernel size from 7 × 7 to 13 × 13 (ablation) | ShuffleMixer · DIV2K validation ×4 | PSNR: 30.28 -> 30.29 | No clear effect |
Closest papers by meaning
- Residual Dense Network for Image Super-Resolution (CVPR 2018 · similarity 0.87)
- Lightweight Image Super-Resolution via Flexible Meta Pruning (ICML 2024 · similarity 0.86)
- WaveMixSR-V2: Enhancing Super-resolution with Higher Efficiency (Student Abstract) (AAAI 2025 · similarity 0.86)
- Efficient Attention-Sharing Information Distillation Transformer for Lightweight Single Image Super-Resolution (AAAI 2025 · similarity 0.86)
- PocketSR: The Super-Resolution Expert in Your Pocket Mobiles (NeurIPS 2025 · similarity 0.85)
- CAMixerSR: Only Details Need More "Attention" (CVPR 2024 · similarity 0.85)
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.