Research map / Architectures and efficiency / Pruning and efficient tuning
GFRRN: Explore the Gaps in Single Image Reflection Removal
Yu Chen, Zewei He, Xingyu Liu, Zixuan Chen, Zhe-Ming Lu
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 |
|---|---|---|---|
| add DAA instead of W-MSA (ablation) | GFRRN · Average494 | PSNR: 26.91 -> 27.33 | Improved |
| add G-AFLB (ablation) | GFRRN · Average494 | PSNR: 27.02 -> 27.33 | Improved |
| add Mona-tuning to pre-trained encoder (ablation) | GFRRN · Average494 | PSNR: 26.70 -> 27.33 | Improved |
| add unified label (ablation) | GFRRN · Average494 | PSNR: 26.96 -> 27.33 | Improved |
| full fine-tuning instead of frozen pre-trained model (ablation) | GFRRN · Average494 | PSNR: 26.70 -> 25.35 | Hurt |
| LoRA instead of frozen model (ablation) | GFRRN · Average494 | PSNR: 26.70 -> 26.51 | Hurt |
| Mona instead of Adaptformer (ablation) | GFRRN · Average494 | PSNR: 27.32 -> 27.33 | Improved |
| Mona instead of Bitfit (ablation) | GFRRN · Average494 | PSNR: 26.86 -> 27.33 | Improved |
| low-frequency unified label instead of I-T (ablation) | GFRRN · Average494 | PSNR: 26.61 -> 27.33 | Improved |
| add unified label (ablation) | DSIT · Average494 | PSNR: 26.50 -> 26.67 | Improved |
| add unified label (ablation) | DSRNet · Average494 | PSNR: 25.72 -> 26.17 | Improved |
| Gaussian mask instead of hard mask in AFLB (ablation) | GFRRN · Average494 | PSNR: 27.08 -> 27.33 | Improved |
Closest papers by meaning
- Single Image Reflection Removal Exploiting Misaligned Training Data and Network Enhancements (CVPR 2019 · similarity 0.89)
- A Lightweight Deep Exclusion Unfolding Network for Single Image Reflection Removal (TPAMI 2025 · similarity 0.88)
- Seeing Deeply and Bidirectionally: A Deep Learning Approach for Single Image Reflection Removal (ECCV 2018 · similarity 0.88)
- LightRR: A Lightweight Network for Single Image Reflection Removal (CVPR 2026 · similarity 0.88)
- FIRM: Flexible Interactive Reflection ReMoval (AAAI 2025 · similarity 0.86)
- Revisiting Single Image Reflection Removal In the Wild (CVPR 2024 · 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.
