FAANet: feature-aligned attention network for real-time multiple object tracking in UAV videos

Multiple object tracking (MOT) in unmanned aerial vehicle (UAV) videos has attracted attention. Because of the observation perspectives UAV, scale changes dramatically and is relatively small. Besides, most MOT algorithms UAV cannot achieve real-time due to tracking-by-detection paradigm. We propose a feature-aligned attention network (FAANet). It mainly consists channel spatial module aggregation module. also improve performance using joint-detection-embedding paradigm structural re-parameterization technique. validate effectiveness with extensive experiments on detection benchmark, achieving new state-of-the-art 44.0 MOTA, 64.6 IDF1 38.24 frames per second running speed single 1080Ti graphics processing unit.