Research map / Image generation / Super-resolution
Memory-Augmented Deep Unfolding Network for Guided Image Super-resolution
Man Zhou, Keyu Yan, Jinshan Pan, Wenqi Ren, Qi Xie, Xiangyong Cao
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 |
|---|---|---|---|
| increase unfolding stages from 1 to 4 (ablation) | MADUNet · GaoFen2 stage study | PSNR: 41.2772 -> 41.8577 | Improved |
| increase unfolding stages from 4 to 5 (ablation) | MADUNet · GaoFen2 stage study | PSNR: 41.8577 -> 41.7545 | Hurt |
| increase unfolding stages from 3 to 4 (ablation) | MADUNet · GaoFen2 stage study | SSIM: 0.9697 -> 0.9697 | No clear effect |
| use memory-augmented unfolding with non-local module versus SRPPNN | MADUNet · WorldView-II | PSNR: 41.4538 -> 41.8577 | Improved |
| use model-driven unfolding versus GPPNN | MADUNet · WorldView-II | PSNR: 41.1622 -> 41.8577 | Improved |
| compare against second-best SRPPNN | MADUNet · GaoFen2 | PSNR: 47.1998 -> 47.2668 | Improved |
| apply pre-trained GaoFen2 model to full-resolution data | MADUNet · GaoFen2 full-resolution | D_s: 0.1139 vs 0.1087 (SFIM) | No clear effect |
| compare against DKN on bicubic depth SR | MADUNet · Depth SR bicubic | average RMSE: 0.09 (4x), 0.36 (8x), 0.85 (16x) lower | Improved |
| compare against DKN on direct depth SR | MADUNet · Depth SR direct | average RMSE: 0.15 (4x), 0.23 (8x), 0.19 (16x) lower | Improved |
| generalize NYU-trained model to Lu dataset at 16x | MADUNet · Lu direct down-sampling | RMSE: 0.03 lower than DKN | No clear effect |
| compare against MINet on MR x4 | MADUNet · MR image SR x4 | PSNR: 35.1998 -> 35.2928 | Improved |
| compare against MINet on MR x2 | MADUNet · MR image SR x2 | PSNR: 38.2553 -> 38.3110 | Improved |
Closest papers by meaning
- Dual-Level Cross-Modality Neural Architecture Search for Guided Image Super-Resolution (TPAMI 2025 · similarity 0.92)
- Deep Learning for Image Super-Resolution: A Survey (TPAMI 2020 · similarity 0.88)
- Dual Graph Regularized Deep Unfolding Network for Guided Depth Map Super-resolution (CVPR 2026 · similarity 0.88)
- Learning Graph Regularisation for Guided Super-Resolution (CVPR 2022 · similarity 0.87)
- Closed-Loop Matters: Dual Regression Networks for Single Image Super-Resolution (CVPR 2020 · similarity 0.87)
- NSARM: Next-Scale Autoregressive Modeling for Robust Real-World Image Super-Resolution (NeurIPS 2026 · similarity 0.86)
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.
