Research map / Detection and segmentation / CNN features
Learning RoI Transformer for Oriented Object Detection in Aerial Images
Check 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 the RoI Transformer to Light-Head R-CNN OBB (ablation) | Light-Head R-CNN OBB · DOTA | mAP: 58.3 -> 63.17 | Helped |
| Add a 2048-d fully connected layer to the RRoI learner (ablation) | RoI Transformer · DOTA | mAP: 63.39 vs. 63.17; 0.22 point drop | No clear effect |
| Enlarge the RRoI context region (ablation) | RoI Transformer · DOTA | mAP: +2.86 points | Helped |
| Remove NMS on transformed RRoIs (ablation) | RoI Transformer · DOTA | mAP: +1.5 points | Helped |
| Use the RoI Transformer instead of deformable PS RoI pooling | Light-Head R-CNN OBB · DOTA | mAP: 63.89 -> 67.74 | Helped |
| Add the RoI Transformer | Light-Head R-CNN OBB with FPN · DOTA | mAP: 66.95 -> 69.56 | Helped |
| Use the RoI Transformer instead of the previous best reported method | Oriented object detection · HRSC2016 | mAP: 84.3 -> 86.2 | Helped |
Closest papers by meaning
- FRED: Towards a Full Rotation-Equivariance in Aerial Image Object Detection (AAAI 2024 · similarity 0.86)
- Semantic-Aware Transformation-Invariant RoI Align (AAAI 2024 · similarity 0.86)
- ABBSPO: Adaptive Bounding Box Scaling and Symmetric Prior based Orientation Prediction for Detecting Aerial Image Objects (CVPR 2025 · similarity 0.86)
- TGBFormer: Transformer-GraphFormer Blender Network for Video Object Detection (AAAI 2025 · similarity 0.85)
- Spatial Transform Decoupling for Oriented Object Detection (AAAI 2024 · similarity 0.85)
- PointOBB: Learning Oriented Object Detection via Single Point Supervision (CVPR 2024 · similarity 0.85)
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.