Abstract
Accurate scale estimation and occlusion handling is a challenging problem in visual tracking. Recently, correlation filter-based trackers have shown impressive results in terms of accuracy, robustness, and speed. However, the model is not robust to scale variation and occlusion. In this paper, we address the problems associated with scale variation and occlusion by employing a scale space filter and multi-block scheme based on a kernelized correlation filter (KCF) tracker. Furthermore, we develop a more robust algorithm using an appearance update model that approximates the change of state of occlusion and deformation. In particular, an adaptive update scheme is presented to make each process robust. The experimental results demonstrate that the proposed method outperformed 29 state-of-the-art trackers on 100 challenging sequences. Specifically, the results obtained with the proposed scheme were improved by 8% and 18% compared to those of the KCF tracker for 49 occlusion and 64 scale variation sequences, respectively. Therefore, the proposed tracker can be a robust and useful tool for object tracking when occlusion and scale variation are involved.
Highlights
This paper proposed simple multi-block-based scale space for kernelized correlation filters
The overall robustness of the system is improved by using an adaptive learning rate for scale space
The overall robustness of the system is improved by using an adaptive learning rate appearance and scale updates with the use of occlusion detection through the distribution of the for appearance and scale updates with the use of occlusion detection through the distribution of the response map
Summary
Visual tracking is a core field of computer vision with many applications such as human computer interaction, surveillance, robotics, driverless vehicles, motion analysis and various intelligent systems.Over the past few decades, visual tracking algorithms with improved performance have been proposed, but they have not provided the desired results in situations involving illumination variation, scale variation, background clutter, and occlusion.The current tracking algorithms mostly use either the generative method [1,2,3,4,5,6,7,8] or the discriminative method [9,10,11,12,13,14]. The correlation filter-based tracker which is discriminative method has been proven to have high efficiency. The correlation filter-based tracker [15,16,17,18,19,20,21,22,23] uses a fixed template size, and it cannot take into account the change in scale. In order to isolate the problem, this paper uses the scale space filter [15] for efficiently estimating the object scale. A part-based method [24,25,26,27,28,29] has been actively researched to solve problems related to changes in the appearance of the target object such as partial occlusion and deformation
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
