48,384 high-resolution aerial images containing 1,041,714 annotated instances across 18 object categories. The dataset focuses on oriented object detection (OBB) and was introduced in the IEEE TPAMI journal for remote sensing research.
Use Cases
- Train rotation-aware object detectors using the OBB (oriented bounding box) coordinates
- Benchmark small object detection performance using the 'small vehicle' and 'helicopter' labels
- Evaluate spatial relationship modeling between categories like 'airport' and 'plane'
Strengths
- 1,041,714 instances annotated with oriented bounding boxes (OBB)
- 18 distinct categories including 'container crane', 'helipad', and 'small vehicle'
- 48,384 images sourced from high-resolution satellite and aerial platforms