Feature Distilled Tracking

Date
2017-12
Language
English
Embargo Lift Date
Committee Members
Degree
Degree Year
Department
Grantor
Journal Title
Journal ISSN
Volume Title
Found At
IEEE
Abstract

Feature extraction and representation is one of the most important components for fast, accurate, and robust visual tracking. Very deep convolutional neural networks (CNNs) provide effective tools for feature extraction with good generalization ability. However, extracting features using very deep CNN models needs high performance hardware due to its large computation complexity, which prohibits its extensions in real-time applications. To alleviate this problem, we aim at obtaining small and fast-to-execute shallow models based on model compression for visual tracking. Specifically, we propose a small feature distilled network (FDN) for tracking by imitating the intermediate representations of a much deeper network. The FDN extracts rich visual features with higher speed than the original deeper network. To further speed-up, we introduce a shift-and-stitch method to reduce the arithmetic operations, while preserving the spatial resolution of the distilled feature maps unchanged. Finally, a scale adaptive discriminative correlation filter is learned on the distilled feature for visual tracking to handle scale variation of the target. Comprehensive experimental results on object tracking benchmark datasets show that the proposed approach achieves 5x speed-up with competitive performance to the state-of-the-art deep trackers.

Description
item.page.description.tableofcontents
item.page.relation.haspart
Cite As
Zhu, G., Wang, J., Wang, P., Wu, Y., & Lu, H. (2017). Feature Distilled Tracking. IEEE Transactions on Cybernetics, 1–13. https://doi.org/10.1109/TCYB.2017.2776977
ISSN
Publisher
Series/Report
Sponsorship
Major
Extent
Identifier
Relation
Journal
Transactions on Cybernetics
Rights
Publisher Policy
Source
Author
Alternative Title
Type
Article
Number
Volume
Conference Dates
Conference Host
Conference Location
Conference Name
Conference Panel
Conference Secretariat Location
Version
Author's manuscript
Full Text Available at
This item is under embargo {{howLong}}