Strong Baseline: Multi-UAV Tracking via YOLOv12 with BoT-SORT-ReID
Yu-Hsi Chen
2025-03-26
Summary
This paper is about creating a simple but effective way to track multiple drones in thermal videos, which are often blurry and hard to see.
What's the problem?
Tracking drones in thermal videos is difficult because the images are low quality, and the drones are small and hard to distinguish from the background.
What's the solution?
The researchers used a combination of existing AI models (YOLOv12 and BoT-SORT) and trained them specifically for this task, without using any fancy techniques to improve the image quality.
Why it matters?
This work matters because it shows that a straightforward approach can be very effective for tracking drones in challenging conditions, providing a baseline for future research.
Abstract
Detecting and tracking multiple unmanned aerial vehicles (UAVs) in thermal infrared video is inherently challenging due to low contrast, environmental noise, and small target sizes. This paper provides a straightforward approach to address multi-UAV tracking in thermal infrared video, leveraging recent advances in detection and tracking. Instead of relying on the YOLOv5 with the DeepSORT pipeline, we present a tracking framework built on YOLOv12 and BoT-SORT, enhanced with tailored training and inference strategies. We evaluate our approach following the metrics from the 4th Anti-UAV Challenge and demonstrate competitive performance. Notably, we achieve strong results without using contrast enhancement or temporal information fusion to enrich UAV features, highlighting our approach as a "Strong Baseline" for the multi-UAV tracking task. We provide implementation details, in-depth experimental analysis, and a discussion of potential improvements. The code is available at https://github.com/wish44165/YOLOv12-BoT-SORT-ReID .