Research map / Detection and segmentation / CNN features
Dynamic Anchor Learning for Arbitrary-Oriented Object Detection
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 DAL label assignment to RetinaNet | RetinaNet · HRSC2016 | mAP: 80.81 -> 88.60 | Helped |
| Add DAL to S^2A-Net | S^2A-Net · DOTA | mAP: 74.12 -> 76.95 | Helped |
| Add DAL label assignment to RetinaNet | RetinaNet · UCAS-AOD | AP75: 39.15 -> 74.30 | Helped |
| Add DAL label assignment to RetinaNet | RetinaNet · ICDAR 2015 | F-measure: 77.5 -> 81.5 | Helped |
| Add multi-scale training and testing (ablation) | Baseline+DAL · ICDAR 2015 | F-measure: 81.5 -> 82.4 | Helped |
| Use output IoU in addition to input IoU, without uncertainty suppression (ablation) | DAL label assignment · HRSC2016 | AP50: 80.8 -> 78.9 | Hurt |
| Add regression uncertainty suppression to combine input and output IoU (ablation) | DAL label assignment · HRSC2016 | AP50: 78.9 -> 85.9 | Helped |
| Add matching-sensitive loss (ablation) | DAL with uncertainty suppression · HRSC2016 | AP75: 57.7 -> 67.6 | Helped |
| Set γ=3 and α=0.9 (ablation) | DAL · HRSC2016 | mAP: 88.6 -> 70.1 | Hurt |
| Use ATSS instead of DAL for label assignment | RetinaNet · HRSC2016 | mAP: 88.6 -> 86.1 | Hurt |
| Add DAL to RetinaNet | RetinaNet · ICDAR 2013 | mAP/F: 77.2 -> 81.3 | Helped |
| Add DAL to RetinaNet | RetinaNet · NWPU VHR-10 | mAP/F: 86.4 -> 88.3 | Helped |
Closest papers by meaning
- Bridging the Gap Between Anchor-Based and Anchor-Free Detection via Adaptive Training Sample Selection (CVPR 2020 · similarity 0.89)
- Region Proposal by Guided Anchoring (CVPR 2019 · similarity 0.88)
- Multiple Anchor Learning for Visual Object Detection (CVPR 2020 · similarity 0.87)
- Improving Single Domain-Generalized Object Detection: A Focus on Diversification and Alignment (CVPR 2024 · similarity 0.86)
- Rethinking Boundary Discontinuity Problem for Oriented Object Detection (CVPR 2024 · similarity 0.84)
- PointOBB-v2: Towards Simpler, Faster, and Stronger Single Point Supervised Oriented Object Detection (ICLR 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.