Research map / Architectures and efficiency / Attention and transformers
GridFormer: Residual Dense Transformer with Grid Structure for Image Restoration in Adverse Weather Conditions
Tao Wang, Kaihao Zhang, Ziqian Shao, Wenhan Luo, Björn Stenger, Tong Lü, Tae‐Kyun Kim, Wei Liu, Hongdong Li
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 feature sampling and channel split in CESA (ablation) | GridFormer · RainDrop | PSNR: 0.89 dB and 1.16 dB improvements | Improved |
| add feature sampling and channel split in CESA (ablation) | GridFormer · SOTS-Indoor | PSNR: 0.92 dB and 2.70 dB improvements | Improved |
| add local enhancement in CESA (ablation) | GridFormer · RainDrop | PSNR: 31.98 -> 32.57 | Improved |
| add local enhancement in CESA (ablation) | GridFormer · SOTS-Indoor | PSNR: 40.51 -> 41.84 | Improved |
| exchange Values instead of Queries between streams (ablation) | GridFormer · RainDrop | PSNR: 31.81 -> 32.57 | Improved |
| pooling strides r=[4,2,2] instead of [2,2,2] (ablation) | GridFormer · SOTS-Indoor | PSNR: 33.81 -> 40.51 | Improved |
| pooling strides r=[4,2,2] instead of [4,4,4] (ablation) | GridFormer · RainDrop | PSNR: 31.62 -> 31.98 | Improved |
| add dense connections in RDTB (ablation) | GridFormer · RainDrop | PSNR: 31.48 -> 32.07 | Improved |
| add local skip connection in RDTB (ablation) | GridFormer · SOTS-Indoor | PSNR: 40.87 -> 41.84 | Improved |
| add local fusion with 1x1 convolution in RDTB (ablation) | GridFormer · RainDrop | PSNR: 32.07 -> 32.05 | No clear effect |
| increase grid columns from c=3 to c=6 at r=3 (ablation) | GridFormer · RainDrop | PSNR: 31.67 -> 32.57 | Improved |
| increase grid rows from r=3 to r=4 (ablation) | GridFormer · RainDrop | PSNR: 32.57 -> 32.05 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- TransWeather — TransWeather: Transformer-Based Restoration of Images Degraded by Adverse Weather Conditions (CVPR 2022) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- All-in-One Transformer for Image Restoration Under Adverse Weather Degradations (TPAMI 2026 · similarity 0.87)
- Restoring Images in Adverse Weather Conditions via Histogram Transformer (ECCV 2024 · similarity 0.86)
- Adapt or Perish: Adaptive Sparse Transformer with Attentive Feature Refinement for Image Restoration (CVPR 2024 · similarity 0.85)
- AWRaCLe: All-Weather Image Restoration Using Visual In-Context Learning (AAAI 2025 · similarity 0.84)
- GridFormer: Point-Grid Transformer for Surface Reconstruction (AAAI 2024 · similarity 0.84)
- MODEM: A Morton-Order Degradation Estimation Mechanism for Adverse Weather Image Recovery (NeurIPS 2025 · similarity 0.84)
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.
