Spatiotemporal Object Detection for Improved Aerial Vehicle Detection in Traffic Monitoring
This work addresses traffic monitoring challenges by enhancing detection accuracy for aerial vehicles, though it is incremental as it builds on existing YOLO-based methods.
The paper tackles vehicle detection from UAV cameras by developing spatiotemporal object detection models, resulting in a 16.22% performance improvement over single-frame models.
This work presents advancements in multi-class vehicle detection using UAV cameras through the development of spatiotemporal object detection models. The study introduces a Spatio-Temporal Vehicle Detection Dataset (STVD) containing 6, 600 annotated sequential frame images captured by UAVs, enabling comprehensive training and evaluation of algorithms for holistic spatiotemporal perception. A YOLO-based object detection algorithm is enhanced to incorporate temporal dynamics, resulting in improved performance over single frame models. The integration of attention mechanisms into spatiotemporal models is shown to further enhance performance. Experimental validation demonstrates significant progress, with the best spatiotemporal model exhibiting a 16.22% improvement over single frame models, while it is demonstrated that attention mechanisms hold the potential for additional performance gains.