DOTA: A Large-scale Dataset for Object Detection in Aerial Images
About
Object detection is an important and challenging problem in computer vision. Although the past decade has witnessed major advances in object detection in natural scenes, such successes have been slow to aerial imagery, not only because of the huge variation in the scale, orientation and shape of the object instances on the earth's surface, but also due to the scarcity of well-annotated datasets of objects in aerial scenes. To advance object detection research in Earth Vision, also known as Earth Observation and Remote Sensing, we introduce a large-scale Dataset for Object deTection in Aerial images (DOTA). To this end, we collect $2806$ aerial images from different sensors and platforms. Each image is of the size about 4000-by-4000 pixels and contains objects exhibiting a wide variety of scales, orientations, and shapes. These DOTA images are then annotated by experts in aerial image interpretation using $15$ common object categories. The fully annotated DOTA images contains $188,282$ instances, each of which is labeled by an arbitrary (8 d.o.f.) quadrilateral To build a baseline for object detection in Earth Vision, we evaluate state-of-the-art object detection algorithms on DOTA. Experiments demonstrate that DOTA well represents real Earth Vision applications and are quite challenging.
Key insights
What aerial detection problem does DOTA address?
DOTA addresses object detection in aerial images, where objects appear at arbitrary orientations and huge scale variations. It provides 2,806 large images with 188,282 instances annotated as oriented quadrilaterals across 15 categories.
How does DOTA differ from prior aerial datasets?
Compared with NWPU VHR-10 and UCAS-AOD, DOTA offers more annotated instances and larger, more varied scenes. Its oriented quadrilateral annotations describe rotated objects that ordinary horizontal boxes capture poorly.
What are DOTA's practical uses?
DOTA serves as a benchmark for developing and evaluating aerial object detection algorithms. The authors note aerial detection helps vehicle counting, remote object tracking, and unmanned driving, though no production deployment is documented.
What limitations and future work do the authors note?
The authors state that detecting densely packed small instances and extremely large instances with arbitrary orientations remains particularly challenging. They plan to continue updating DOTA to grow in size and scope.