Abstract
Discriminative Correlation Filters (DCF) based trackers have achieved remarkable performance in visual object tracking in recent years. The trackers represent the target with hand-craft features or deep features. To reduce the computational cost and irrelevant information, such trackers choose the region of interest (ROI) as a search window to search the target rather than search exhaustively in each frame. The size of search window, which is fixed in most trackers, can greatly affect the performance of tracker. In order to characterize the motion state of the target, this paper presents the velocity and acceleration in the field of visual object tracking. It is observed that the acceleration can change the search window dynamically and achieve better performance. Experiments on the popular datasets OTB50 and OTB100 demonstrate that the DCF based trackers including the state-of-the-art (SOTA) trackers improve their performance by exploiting acceleration.
Highlights
Visual object tracking is an important branch in computer vision for its wide applications in surveillance, robot vision, positioning technology among others
Since popular datasets and benchmarks such as OTB-50 [1], OTB-100 [2], as well as challenges for visual object tracking such as VOT2016 [3], MOT [4], have been collected, many accurate and efficient tracking algorithms spring up. Most of those trackers are based on discriminative correlation filters (DCF) and deep learning (DL)
Sparked by the impressive performance of Minimum Output Sum of Squared Error (MOSSE), Henriques et al [6] exploited the circulant structure of dense sampling with kernels (CSK), which demonstrates the
Summary
Visual object tracking is an important branch in computer vision for its wide applications in surveillance, robot vision, positioning technology among others. Since popular datasets and benchmarks such as OTB-50 [1], OTB-100 [2], as well as challenges for visual object tracking such as VOT2016 [3], MOT [4], have been collected, many accurate and efficient tracking algorithms spring up Most of those trackers are based on discriminative correlation filters (DCF) and deep learning (DL). Bhat et al [11] proposed a tracker using the surroundings to identify the target This tracker is evaluated on five tracking benchmarks and sets a new SOTA on three of these benchmarks. Trackers such as KCF and C-COT ignore the potential information of the target motion. All trackers selected at random, including SOTA trackers, improve their performance by applying the method proposed in this paper
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
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 callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
