SDG-Track: A Heterogeneous Observer-Follower Framework for High-Resolution UAV Tracking on Embedded Platforms

TL;DR

SDG-Track is a novel UAV tracking framework that balances high-resolution detection accuracy with real-time processing on embedded platforms by combining a low-frequency high-capacity detector with high-frequency optical flow tracking.

Contribution

It introduces an Observer-Follower architecture with a dual-space recovery mechanism for robust, real-time UAV tracking on resource-limited embedded systems.

Findings

Achieves 35.1 FPS throughput on NVIDIA Jetson Orin Nano.

Retains 97.2% detection precision in real-world conditions.

Successfully tracks agile FPV drones in operational scenarios.

Abstract

Real-time tracking of small unmanned aerial vehicles (UAVs) on edge devices faces a fundamental resolution-speed conflict. Downsampling high-resolution imagery to standard detector input sizes causes small target features to collapse below detectable thresholds. Yet processing native 1080p frames on resource-constrained platforms yields insufficient throughput for smooth gimbal control. We propose SDG-Track, a Sparse Detection-Guided Tracker that adopts an Observer-Follower architecture to reconcile this conflict. The Observer stream runs a high-capacity detector at low frequency on the GPU to provide accurate position anchors from 1920x1080 frames. The Follower stream performs high-frequency trajectory interpolation via ROI-constrained sparse optical flow on the CPU. To handle tracking failures from occlusion or model drift caused by spectrally similar distractors, we introduce Dual-Space Recovery, a training-free re-acquisition mechanism combining color histogram matching with geometric consistency constraints. Experiments on a ground-to-air tracking station demonstrate that SDG-Track achieves 35.1 FPS system throughput while retaining 97.2\% of the frame-by-frame detection precision. The system successfully tracks agile FPV drones under real-world operational conditions on an NVIDIA Jetson Orin Nano. Our paper code is publicly available at https://github.com/Jeffry-wen/SDG-Track