Share Email Print

Proceedings Paper

Incrementally update CNN for visual tracking using active learning and artificial data
Author(s): Yunqiu Lv; Kai Liu; Fei Cheng; Wei Li
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

Deep learning has been widely used in visual tracking due to strong feature extraction ability of convolutional neural network(CNN). Many trackers pre-train CNN primarily and fine-tune it during tracking, which could improve representation ability from off-line database and adjust to appearance variation of the interested object. However, since target information is limited, the network is likely to overfit to a single target state. In this paper, an update strategy composed of two modules is proposed. First, we fine-tune the pre-trained CNN using active learning that emphasizes the most discriminative data iteratively. Second, artificial convolutional features generated from empirical distribution are employed to train fully connected layers, which makes up the deficiency of training examples. Experiments evaluated on VOT2016 benchmark shows that our algorithm outperforms many state-of-the-art trackers.

Paper Details

Date Published: 29 October 2018
PDF: 5 pages
Proc. SPIE 10836, 2018 International Conference on Image and Video Processing, and Artificial Intelligence, 1083610 (29 October 2018); doi: 10.1117/12.2504317
Show Author Affiliations
Yunqiu Lv, Xidian Univ. (China)
Kai Liu, Xidian Univ. (China)
Fei Cheng, Xidian Univ. (China)
Wei Li, Xidian Univ. (China)

Published in SPIE Proceedings Vol. 10836:
2018 International Conference on Image and Video Processing, and Artificial Intelligence
Ruidan Su, Editor(s)

© SPIE. Terms of Use
Back to Top