Research map / Vision-language and multimodal / CLIP and zero-shot
RAW-Adapter: Adapting Pre-Trained Visual Model to Camera RAW Images and a Benchmark
Ziteng Cui, Jianfei Yang, Tatsuya Harada
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 RAW-based data augmentation (ablation) | RAW-Adapter · PASCAL RAW-D OOD Low-light | mAP: 75.9 -> 84.9 | Improved |
| add RAW-based data augmentation (ablation) | RAW-Adapter · PASCAL RAW-D OOD Normal | mAP: 88.7 -> 88.9 | Improved |
| add RAW-based data augmentation (ablation) | Baseline (direct RAW) · PASCAL RAW-D OOD Normal | mAP: 87.7 -> 87.1 | Hurt |
| add RAW-based data augmentation (ablation) | Dirty-Pixel · PASCAL RAW-D OOD Defocus Blur | mAP: 32.5 -> 50.4 | Improved |
| add RAW-based data augmentation (ablation) | RAW-Adapter · PASCAL RAW-D OOD Defocus Blur | mAP: 32.3 -> 64.8 | Improved |
| add RAW-based data augmentation (ablation) | RAW-Adapter · LOD (RetinaNet) | mAP: 62.1 -> 62.8 | Improved |
| add RAW-based data augmentation (ablation) | Reconfig-ISP · LOD (Sparse-RCNN) | mAP: 58.9 -> 58.9 | No clear effect |
| add RAW-based data augmentation (ablation) | RAW-Adapter · iPhone XS Max-D OOD Low-light | mIoU: 29.79 -> 49.99 | Improved |
| add RAW-based data augmentation (ablation) | RAW-Adapter · iPhone XS Max-D OOD Rain | mIoU: 5.53 -> 21.02 | Improved |
| add RAW-based data augmentation (ablation) | RAW-Adapter · iPhone XS Max-D OOD Moire | mIoU: 37.17 -> 31.22 | Hurt |
| add RAW-based data augmentation (ablation) | Baseline (direct RAW) · iPhone XS Max-D OOD Normal | mIoU: 57.55 -> 56.46 | Hurt |
| add RAW-based data augmentation (ablation) | Reconfig-ISP · PASCAL RAW-D OOD Moire | mAP: 80.3 -> 78.3 | Hurt |
Closest papers by meaning
- RAW-Adapter: Adapting Pretrained Visual Model to Camera RAW Images (ECCV 2024 · similarity 0.96)
- ReRAW: RGB-to-RAW Image Reconstruction via Stratified Sampling for Efficient Object Detection on the Edge (CVPR 2025 · similarity 0.86)
- RAWild: Toward Sensor-Agnostic RAW Object Detection via Physics-Guided Curve and Grid Modeling (NeurIPS 2026 · similarity 0.85)
- Guiding a Harsh-Environments Robust Detector via RAW Data Characteristic Mining (AAAI 2024 · similarity 0.85)
- Linear Differential Vision Transformer: Learning Visual Contrasts via Pairwise Differentials (NeurIPS 2025 · similarity 0.84)
- Compress Clean Signal from Noisy Raw Image: A Self-Supervised Approach (ICML 2024 · 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.
