Abstract
The visual object tracking problem seeks to track an arbitrary object in a video, and many deep convolutional neural network-based algorithms have achieved significant performance improvements in recent years. However, most of them do not guarantee real-time operation due to the large computation overhead for deep feature extraction. This paper presents a single-crop visual object tracking algorithm based on a fully convolutional Siamese network (SiamFC). The proposed algorithm significantly reduces the computation burden by extracting multiple scale feature maps from a single image crop. Experimental results show that the proposed algorithm demonstrates superior speed performance in comparison with that of SiamFC.
Highlights
The objective of visual tracking is to track an arbitrary object in a video, where the ground-truth bounding box is given in the first frame [1,2,3]
The tracking algorithms based on the correlation filter train a regressor from a set of training samples by updating the weights of filters, and the operation is performed in the Fourier domain using the fast Fourier transform [7,8,9]
Since the correlation filters must be updated during tracking and the training of the convolutional neural networks requires heavy computation, they suffer from speed degeneration
Summary
The objective of visual tracking is to track an arbitrary object in a video, where the ground-truth bounding box is given in the first frame [1,2,3]. Since the correlation filters must be updated during tracking and the training of the convolutional neural networks requires heavy computation, they suffer from speed degeneration. SiamFC requires multi-scale tests for scale estimation such that image patches with different scales per each image frame must be tested independently This causes an increase of computation time proportional to the number of image patches. The proposed approach uses only a single-cropped patch per each image frame to generate the feature maps of multiple scales with less computation. This significantly improves tracking speed of SiamFC with a negligible precision loss
Sign-in/Register to view highlights & summary 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.
