Research map / Image generation / Super-resolution
Image Super-Resolution via Dual-State Recurrent Networks
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 |
|---|---|---|---|
| add dual LR and HR recurrent states rather than using a single state (ablation) | DSRN · NTIRE SR 2017 track 1, ×2/×3/×4 | PSNR: 34.66 -> 34.89; 30.80 -> 30.95; 28.80 -> 28.99 | Helped |
| add delayed HR-to-LR feedback (ablation) | DSRN · NTIRE SR 2017 track 1, ×2/×3/×4 | PSNR: 34.89 -> 34.96; 30.95 -> 31.12; 28.99 -> 29.03 | Helped |
| add parameter sharing across recurrent steps (ablation) | DSRN · NTIRE SR 2017 track 1, ×2/×3/×4 | PSNR: 34.71 -> 34.96; 30.85 -> 31.12; 28.81 -> 29.03 | Helped |
| increase unrolling length T (ablation) | DSRN · Set5 ×2 | PSNR | Helped |
| randomly sample unrolling length T during training with stochastic depth (ablation) | DSRN · validation accuracy | accuracy | No clear effect |
| average predictions from all unrolled states rather than use only the last state (ablation) | DSRN · image super-resolution | restoration performance | Helped |
| use DSRN rather than DRRN | image super-resolution · Set14 ×3 | PSNR: 29.96 -> 30.26 | Helped |
| use DSRN rather than DRRN | image super-resolution · Set5 ×2 | PSNR: 37.74 -> 37.66 | Hurt |
Closest papers by meaning
- Feedback Network for Image Super-Resolution (CVPR 2019 · similarity 0.91)
- Frame-Recurrent Video Super-Resolution (CVPR 2018 · similarity 0.91)
- Closed-Loop Matters: Dual Regression Networks for Single Image Super-Resolution (CVPR 2020 · similarity 0.90)
- Recurrent Back-Projection Network for Video Super-Resolution (CVPR 2019 · similarity 0.90)
- Exploring the Low-Pass Filtering Behavior in Image Super-Resolution (ICML 2024 · similarity 0.87)
- ResDiff: Combining CNN and Diffusion Model for Image Super-resolution (AAAI 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.