Abstract
Discriminative Correlation Filter (DCF) based trackers are quite efficient in tracking objects by exploiting the circulant structure. The kernel trick further improves the performance of such trackers. The unwanted boundary effects, however, are difficult to solve in the kernelized correlation models. In this paper, we propose a novel Constrained Multi-Kernel Correlation tracking Filter (CMKCF), which applies spatial constraints to address this drawback. We build the multi-kernel models for multi-channel features with three different attributes, and then employ a spatial cropping operator on the semi-kernel matrix to address the boundary effects. For the constrained optimization solution, we develop an Alternating Direction Method of Multipliers (ADMM) based algorithm to learn our multi-kernel filters efficiently in the frequency domain. In particular, we suggest an adaptive updating mechanism by exploiting the feedback from high-confidence tracking results to avoid corruption in the model. Extensive experimental results demonstrate that the proposed method performs favorably on OTB-2013, OTB-2015, VOT-2016 and VOT-2018 dataset against several state-of-the-art methods.
Highlights
V ISUAL tracking has a plethora of practical applications in computer vision, including robotics [1], surveillance [2], video processing and biological image analysis [3]
Success metric measures the percentage of frames where the Intersection over Union (IoU) ratios of predicted and groundtruth bounding boxes are larger than a given threshold and the overall success performance is indicated by the Area Under the Curve (AUC) of success plots for all the thresholds
Considering clarity and representation, we mainly focus on the performance of the most common challenge factors, namely Occlusion (OCC), In-Plane or Out-of-Plane Rotation (IPR or OPR), Fast Motion (FM), Scale Variation (SV) and Background Clutter (BC)
Summary
V ISUAL tracking has a plethora of practical applications in computer vision, including robotics [1], surveillance [2], video processing and biological image analysis [3]. For the constrained optimization solution, we develop an efficient Alternating Direction Method of Multipliers (ADMM) [27] based algorithm to learn our multi-kernel CFs. we suggest an adaptive updating mechanism to avoid the model corruption problem. In general DCF tracking approaches (e.g. KCF [8] and ECO [28]), they utilize a certain learning rate to update the training features in every frame or every several frames to make the model more adaptable This may work in the scenes of a very short-term loss of a target. The Constrained Multi-Kernel Correlation Filter (CMKCF) is elaborated in detail in Section III (including learning CMKCFs, optimization algorithm, scale estimation and occlusion-aware update).
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.
