The UAV Single Target Tracking Algorithm Based on Multimodal Data

With the rapid development of drone technology, drone target tracking has become one of the key technologies in drone applications. However, single modal data often has limitations in complex scenes, such as changes in lighting, occlusion, and background interference. Therefore, multimodal data fusion has shown great potential in the field of target tracking. Based on the TCTrack algorithm, this paper conducted multimodal data fusion feature extraction, combined with visible light, thermal infrared information, and TCTrack's utilization of contextual information between consecutive frames, significantly improving the robustness and accuracy of target tracking. The experimental results show that the algorithm exhibits excellent tracking performance in various complex scenarios, significantly improves target localization accuracy, and demonstrates strong adaptability in dynamic environments.