EnKCF: Ensemble of Kernelized Correlation Filters for High-Speed Object Tracking

TL;DR

EnKCF is a high-speed ensemble tracking algorithm combining multiple kernelized correlation filters with a particle filter, achieving superior accuracy and speed on benchmark datasets for real-time embedded applications.

Contribution

The paper introduces EnKCF, an ensemble of KCFs with a particle filter for high-speed, accurate object tracking without offline training, optimized for embedded systems.

Findings

Achieved 340 fps on OTB100 and 416 fps on UAV123 datasets.

Outperformed other high-speed trackers by over 5% in precision.

Demonstrated better accuracy and speed than DCF and KCF methods.

Abstract

Computer vision technologies are very attractive for practical applications running on embedded systems. For such an application, it is desirable for the deployed algorithms to run in high-speed and require no offline training. To develop a single-target tracking algorithm with these properties, we propose an ensemble of the kernelized correlation filters (KCF), we call it EnKCF. A committee of KCFs is specifically designed to address the variations in scale and translation of moving objects. To guarantee a high-speed run-time performance, we deploy each of KCFs in turn, instead of applying multiple KCFs to each frame. To minimize any potential drifts between individual KCFs transition, we developed a particle filter. Experimental results showed that the performance of ours is, on average, 70.10% for precision at 20 pixels, 53.00% for success rate for the OTB100 data, and 54.50% and 40.2% for the UAV123 data. Experimental results showed that our method is better than other high-speed trackers over 5% on precision on 20 pixels and 10-20% on AUC on average. Moreover, our implementation ran at 340 fps for the OTB100 and at 416 fps for the UAV123 dataset that is faster than DCF (292 fps) for the OTB100 and KCF (292 fps) for the UAV123. To increase flexibility of the proposed EnKCF running on various platforms, we also explored different levels of deep convolutional features.