Two-stage learning to robust visual track via CNNs

Dan Hu, Xingshe Zhou, Xiaohao Yu, Zhiqiang Hou

科研成果: 书/报告/会议事项章节会议稿件同行评审

1 引用 (Scopus)

摘要

Convolutional Neural Networks (CNN) are an alternative type of deep neural network that can be used to model local correlations and reduce translation variations, which have demonstrated great performance in some computer vision areas except the visual tracking due to the lack of training data. In this paper, we explore applying a two-stage learning CNN as a generic feature extractor offline pretrained with a large auxiliary dataset and then transfer its rich feature hierarchies to the robust visual tracking task. Instead of traditional neuron models in CNNs, we introduce a strategy to use ReLU for training acceleration. Empirical comparisons prove our CNN based tracker outperforms several state-of-the-art methods on an open tracking benchmark.

源语言英语
主期刊名Image and Graphics - 8th International Conference, ICIG 2015, Proceedings
编辑Yu-Jin Zhang
出版商Springer Verlag
491-498
页数8
ISBN(印刷版)9783319219684
DOI
出版状态已出版 - 2015
活动8th International Conference on Image and Graphics, ICIG 2015 - Tianjin, 中国
期限: 13 8月 201516 8月 2015

出版系列

姓名Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
9219
ISSN(印刷版)0302-9743
ISSN(电子版)1611-3349

会议

会议8th International Conference on Image and Graphics, ICIG 2015
国家/地区中国
Tianjin
时期13/08/1516/08/15

指纹

探究 'Two-stage learning to robust visual track via CNNs' 的科研主题。它们共同构成独一无二的指纹。

引用此