GL-DT: Multi-UAV Detection and Tracking with Global-Local Integration

2025-10-10

AI Summary

In complex scenarios, UAV-based multi-object tracking suffers from trajectory discontinuity due to small object scales and frequent occlusions. To address this, we propose GL-DT, a global-local fusion framework. Methodologically: (1) We design the Spatio-Temporal Feature Fusion (STFF) module to jointly model motion and appearance cues, significantly improving detection robustness for small targets; (2) We introduce JPTrack, a data association algorithm that jointly optimizes motion prediction and re-identification to suppress ID switches and track fragmentation. Experiments demonstrate that GL-DT achieves state-of-the-art performance in both MOTA and IDF1 while maintaining real-time inference speed, thereby substantially enhancing trajectory continuity and stability.

Technology Category

• Computer Vision: Motion & Tracking
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• Web Mining and Content Analysis: Topic discovery and tracking
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Security and Privacy: Tracking, profiling, and countermeasures against them

Abstract

The extensive application of unmanned aerial vehicles (UAVs) in military reconnaissance, environmental monitoring, and related domains has created an urgent need for accurate and efficient multi-object tracking (MOT) technologies, which are also essential for UAV situational awareness. However, complex backgrounds, small-scale targets, and frequent occlusions and interactions continue to challenge existing methods in terms of detection accuracy and trajectory continuity. To address these issues, this paper proposes the Global-Local Detection and Tracking (GL-DT) framework. It employs a Spatio-Temporal Feature Fusion (STFF) module to jointly model motion and appearance features, combined with a global-local collaborative detection strategy, effectively enhancing small-target detection. Building upon this, the JPTrack tracking algorithm is introduced to mitigate common issues such as ID switches and trajectory fragmentation. Experimental results demonstrate that the proposed approach significantly improves the continuity and stability of MOT while maintaining real-time performance, providing strong support for the advancement of UAV detection and tracking technologies.

Problem

• Improving multi-UAV detection accuracy in complex backgrounds
• Addressing trajectory fragmentation and ID switches in tracking
• Enhancing small-target detection with global-local collaboration

Innovation

• Spatio-temporal fusion for motion and appearance modeling
• Global-local collaborative detection for small targets
• JPTrack algorithm reduces ID switches and fragmentation