DEFEATnet-A Deep Conventional Image Representation for Image Classification

doi:10.1109/TCSVT.2015.2389413

	DEFEATnet-A Deep Conventional Image Representation for Image Classification
	Shenghua Gao1 ; Lixin Duan 2; Ivor W. Tsang 3
	2016-03
发表期刊	IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY (IF:8.3[JCR-2023],7.1[5-Year])
ISSN	1051-8215
卷号	26 期号:3 页码:494-505
发表状态	已发表
DOI	10.1109/TCSVT.2015.2389413
摘要	To study underlying possibilities for the successes of conventional image representation and deep neural networks (DNNs) in image representation, we propose a deep feature extraction, encoding, and pooling network (DEFEATnet) architecture, which is a marriage between conventional image representation approaches and DNNs. In particular, in DEFEATnet, each layer consists of three components: feature extraction, feature encoding, and pooling. The primary advantage of DEFEATnet is twofold. First, it consolidates the prior knowledge (e.g., translation invariance) from extracting, encoding, and pooling handcrafted features, as in the conventional feature representation approaches. Second, it represents the object parts at different granularities by gradually increasing the local receptive fields in different layers, as in DNNs. Moreover, DEFEATnet is a generalized framework that can readily incorporate all types of local features as well as all kinds of well-designed feature encoding and pooling methods. Since prior knowledge is preserved in DEFEATnet, it is especially useful for image representation on small/medium size data sets, where DNNs usually fail due to the lack of sufficient training data. Promising experimental results clearly show that DEFEATnets outperform shallow conventional image representation approaches by a large margin when the same type of features, feature encoding and pooling are used. The extensive experiments also demonstrate the effectiveness of the deep architecture of our DEFEATnet in improving the robustness for image presentation.
关键词	Conventional image representation deep architecture feature encoding local max pooling
URL	查看原文
收录类别	SCI ; EI
语种	英语
资助项目	National Science Foundation of China[61502304]
WOS研究方向	Engineering
WOS类目	Engineering, Electrical & Electronic
WOS记录号	WOS:000372547400006
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
EI入藏号	20161702307652
EI主题词	Encoding (symbols) ; Extraction ; Feature extraction ; Network architecture
EI分类号	Data Processing and Image Processing:723.2 ; Chemical Operations:802.3
WOS关键词	ALGORITHM ; FEATURES
原始文献类型	Article
来源库	IEEE
引用统计	正在获取...
文献类型	期刊论文
条目标识符	https://kms.shanghaitech.edu.cn/handle/2MSLDSTB/1911
专题	信息科学与技术学院_PI研究组_高盛华组
作者单位	1.ShanghaiTech University, Shanghai, China 2.Amazon, Seattle, WA, USA 3.University of Technology, Sydney, NSW, Australia
第一作者单位	上海科技大学
第一作者的第一单位	上海科技大学
推荐引用方式 GB/T 7714	Shenghua Gao,Lixin Duan,Ivor W. Tsang. DEFEATnet-A Deep Conventional Image Representation for Image Classification[J]. IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY,2016,26(3):494-505.
APA	Shenghua Gao,Lixin Duan,&Ivor W. Tsang.(2016).DEFEATnet-A Deep Conventional Image Representation for Image Classification.IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY,26(3),494-505.
MLA	Shenghua Gao,et al."DEFEATnet-A Deep Conventional Image Representation for Image Classification".IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY 26.3(2016):494-505.