Matryoshka Peek: Toward Learning Fine-Grained, Robust, Discriminative Features for Product Search

doi:10.1109/TMM.2017.2655422

	Matryoshka Peek: Toward Learning Fine-Grained, Robust, Discriminative Features for Product Search
	Kyaw, Zawlin 1; Qi, Shuhan 2; Gao, Ke 3; Zhang, Hanwang 1; Zhang, Luming 4; Xiao, Jun 5; Wang, Xuan 6; Chua, Tat-Seng 1
	2017-06-01
发表期刊	IEEE TRANSACTIONS ON MULTIMEDIA
ISSN	1520-9210
卷号	19 期号:6 页码:1272-1284
摘要	In sharp contrast to the traditional category/subcategory level image retrieval, product image search aims to find the images containing the exact same product. This is a challenging problem because in addition to being robust under different imaging conditions such as varying viewpoints and illumination changes, the features should also be able to distinguish the specific product among many similar products. Consequently, it is important to utilize a large dataset, containing many product classes, to learn a strongly discriminative representation. Building such a dataset requires laborious manual annotation. Toward learning fine-grained, robust, discriminative features for product image search, we present a novel paradigm that can construct the required dataset without any human annotation. Unlike other fine-grained recognition works that rely on high-quality annotated datasets and are very narrowly focused on a specific object category, our method handles multiple object classes and requires minimum human effort. First, an ImageNet pretrained model is used to generate product clusters. As the original features from ImageNet are not discriminative, the clusters generated by this unsupervised procedure contain much noise. We alleviate noise by explicitly modeling noise distribution and automatically detecting errors during learning. The proposed paradigm is general, requires minimum human efforts, and is applicable to any deep learning task where fine-grained discriminative features are desired. Extensive experiments on the ALISC dataset have demonstrated that our approach is sound and effective, surpassing the baseline GoogleNet model by 15.09%.
关键词	Feature extraction image representation robust learning image retrieval
DOI	10.1109/TMM.2017.2655422
收录类别	SCI
语种	英语
资助项目	National Research Foundation, Prime Minister's Office, Singapore, under its IRC@SG Funding Initiative ; National Nature Science Foundation of China[61525206] ; National Nature Science Foundation of China[61271428] ; National Nature Science Foundation of China[61572169] ; National Nature Science Foundation of China[61472266] ; International Exchange and Cooperation Foundation of Shenzhen City[GJHZ20150312114149569] ; National University of Singapore (Suzhou) Research Institute ; Fundamental Research Funds for the Central Universities
WOS研究方向	Computer Science ; Telecommunications
WOS类目	Computer Science, Information Systems ; Computer Science, Software Engineering ; Telecommunications
WOS记录号	WOS:000404059400013
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
引用统计	被引频次：3[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/7072
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Qi, Shuhan
作者单位	1.Natl Univ Singapore, Sch Comp, Singapore 119077, Singapore 2.Harbin Inst Technol, ShenZhen Grad Sch, Shenzhen 518055, Peoples R China 3.Chinese Acad Sci, Inst Comp Technol, Beijing 100190, Peoples R China 4.Hefei Univ Technol, Hefei 132312, Peoples R China 5.Zhejiang Univ, Hangzhou 132312, Zhejiang, Peoples R China 6.Harbin Inst Technol, Comp Applicat Res Ctr, ShenZhen Grad Sch, Shenzhen 518055, Peoples R China
推荐引用方式 GB/T 7714	Kyaw, Zawlin,Qi, Shuhan,Gao, Ke,et al. Matryoshka Peek: Toward Learning Fine-Grained, Robust, Discriminative Features for Product Search[J]. IEEE TRANSACTIONS ON MULTIMEDIA,2017,19(6):1272-1284.
APA	Kyaw, Zawlin.,Qi, Shuhan.,Gao, Ke.,Zhang, Hanwang.,Zhang, Luming.,...&Chua, Tat-Seng.(2017).Matryoshka Peek: Toward Learning Fine-Grained, Robust, Discriminative Features for Product Search.IEEE TRANSACTIONS ON MULTIMEDIA,19(6),1272-1284.
MLA	Kyaw, Zawlin,et al."Matryoshka Peek: Toward Learning Fine-Grained, Robust, Discriminative Features for Product Search".IEEE TRANSACTIONS ON MULTIMEDIA 19.6(2017):1272-1284.