AU-AIR: A Multi-modal Unmanned Aerial Vehicle Dataset for Low Altitude Traffic Surveillance

Unmanned aerial vehicles (UAVs) with mounted cameras have the advantage of\ncapturing aerial (bird-view) images. The availability of aerial visual data and\nthe recent advances in object detection algorithms led the computer vision\ncommunity to focus on object detection tasks on aerial images. As a result of\nthis, several aerial datasets have been introduced, including visual data with\nobject annotations. UAVs are used solely as flying-cameras in these datasets,\ndiscarding different data types regarding the flight (e.g., time, location,\ninternal sensors). In this work, we propose a multi-purpose aerial dataset\n(AU-AIR) that has multi-modal sensor data (i.e., visual, time, location,\naltitude, IMU, velocity) collected in real-world outdoor environments. The\nAU-AIR dataset includes meta-data for extracted frames (i.e., bounding box\nannotations for traffic-related object category) from recorded RGB videos.\nMoreover, we emphasize the differences between natural and aerial images in the\ncontext of object detection task. For this end, we train and test mobile object\ndetectors (including YOLOv3-Tiny and MobileNetv2-SSDLite) on the AU-AIR\ndataset, which are applicable for real-time object detection using on-board\ncomputers with UAVs. Since our dataset has diversity in recorded data types, it\ncontributes to filling the gap between computer vision and robotics. The\ndataset is available at https://bozcani.github.io/auairdataset.\n