Next Article in Journal
A Reciprocal-Selection-Based ‘Win–Win’ Overlay Spectrum-Sharing Scheme for Device-to-Device-Enabled Cellular Network
Previous Article in Journal
Understanding and Enhancement of Internal Clustering Validation Indexes for Categorical Data
Article Menu

Export Article

Open AccessArticle
Algorithms 2018, 11(11), 178;

Deep Directional Network for Object Tracking

1,2,* and 1
School of Electronic & Information Engineering, Nanjing University of Information Science & Technology, Nanjing 210044, China
Jiangsu Collaborative Innovation Center on Atmospheric Environment and Equipment Technology, Nanjing University of Information Science & Technology, Nanjing 210044, China
Author to whom correspondence should be addressed.
Received: 10 October 2018 / Revised: 28 October 2018 / Accepted: 1 November 2018 / Published: 5 November 2018
Full-Text   |   PDF [3003 KB, uploaded 6 November 2018]   |  


Existing object trackers are mostly based on correlation filtering and neural network frameworks. Correlation filtering is fast but has poor accuracy. Although a neural network can achieve high precision, a large amount of computation increases the tracking time. To address this problem, we utilize a convolutional neural network (CNN) to learn object direction. We propose a target direction classification network based on CNNs that has a directional shortcut to the tracking target, unlike the particle filter that randomly finds the target. Our network uses an end-to-end approach to determine scale variation that has good robustness to scale variation sequences. In the pretraining stage, the Visual Object Tracking Challenges (VOT) dataset is used to train the network for learning positive and negative sample classification and direction classification. In the online tracking stage, the sliding window operation is performed by using the obtained directional information to determine the exact position of the object. The network only calculates a single sample, which guarantees a low computational burden. The positive and negative sample redetection strategies can successfully ensure that the samples are not lost. The one-pass evaluation (OPE) evaluation results of the object tracking benchmark (OTB) demonstrate that the algorithm is very robust and is also faster than several deep trackers. View Full-Text
Keywords: object tracking; deep directional network; scale variation; sliding window; single sample object tracking; deep directional network; scale variation; sliding window; single sample

Figure 1

This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).

Share & Cite This Article

MDPI and ACS Style

Hu, Z.; Shi, X. Deep Directional Network for Object Tracking. Algorithms 2018, 11, 178.

Show more citation formats Show less citations formats

Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Related Articles

Article Metrics

Article Access Statistics



[Return to top]
Algorithms EISSN 1999-4893 Published by MDPI AG, Basel, Switzerland RSS E-Mail Table of Contents Alert
Back to Top