Robust Image Analysis With Sparse Representation on Quantized Visual Features

CORC > 自动化研究所 > 中国科学院自动化研究所 > 模式识别国家重点实验室 > 多媒体计算与图形学团队

	Robust Image Analysis With Sparse Representation on Quantized Visual Features
	Bao, Bing-Kun1,2 ; Zhu, Guangyu 3; Shen, Jialie 4; Yan, Shuicheng 5
刊名	IEEE TRANSACTIONS ON IMAGE PROCESSING
	2013-03-01
卷号	22 期号:3 页码:860-871
关键词	Image classification quantized visual feature sparse representation
英文摘要	Recent techniques based on sparse representation (SR) have demonstrated promising performance in high-level visual recognition, exemplified by the highly accurate face recognition under occlusion and other sparse corruptions. Most research in this area has focused on classification algorithms using raw image pixels, and very few have been proposed to utilize the quantized visual features, such as the popular bag-of-words feature abstraction. In such cases, besides the inherent quantization errors, ambiguity associated with visual word assignment and misdetection of feature points, due to factors such as visual occlusions and noises, constitutes the major cause of dense corruptions of the quantized representation. The dense corruptions can jeopardize the decision process by distorting the patterns of the sparse reconstruction coefficients. In this paper, we aim to eliminate the corruptions and achieve robust image analysis with SR. Toward this goal, we introduce two transfer processes (ambiguity transfer and mis-detection transfer) to account for the two major sources of corruption as discussed. By reasonably assuming the rarity of the two kinds of distortion processes, we augment the original SR-based reconstruction objective with l(0)-norm regularization on the transfer terms to encourage sparsity and, hence, discourage dense distortion/transfer. Computationally, we relax the nonconvex l(0)-norm optimization into a convex l(1)-norm optimization problem, and employ the accelerated proximal gradient method to optimize the convergence provable updating procedure. Extensive experiments on four benchmark datasets, Caltech-101, Caltech-256, Corel-5k, and CMU pose, illumination, and expression, manifest the necessity of removing the quantization corruptions and the various advantages of the proposed framework.
WOS标题词	Science & Technology ; Technology
类目[WOS]	Computer Science, Artificial Intelligence ; Engineering, Electrical & Electronic
研究领域[WOS]	Computer Science ; Engineering
关键词[WOS]	LOCAL BINARY PATTERNS ; FACE RECOGNITION ; CLASSIFICATION ; REGRESSION ; SHRINKAGE ; ALGORITHM ; SELECTION ; SYSTEMS ; SCALE ; LASSO
收录类别	SCI
语种	英语
WOS记录号	WOS:000318014300002
公开日期	2015-09-22
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/8861]
专题	自动化研究所_模式识别国家重点实验室_多媒体计算与图形学团队
作者单位	1.Chinese Acad Sci, Natl Lab Pattern Recognit, Inst Automat, Beijing 100190, Peoples R China 2.China Singapore Inst Digital Media, Singapore 119613, Singapore 3.Univ Calif Berkeley, Berkeley, CA 94720 USA 4.Singapore Management Univ, Singapore 188065, Singapore 5.Natl Univ Singapore, Singapore 117576, Singapore
推荐引用方式 GB/T 7714	Bao, Bing-Kun,Zhu, Guangyu,Shen, Jialie,et al. Robust Image Analysis With Sparse Representation on Quantized Visual Features[J]. IEEE TRANSACTIONS ON IMAGE PROCESSING,2013,22(3):860-871.
APA	Bao, Bing-Kun,Zhu, Guangyu,Shen, Jialie,&Yan, Shuicheng.(2013).Robust Image Analysis With Sparse Representation on Quantized Visual Features.IEEE TRANSACTIONS ON IMAGE PROCESSING,22(3),860-871.
MLA	Bao, Bing-Kun,et al."Robust Image Analysis With Sparse Representation on Quantized Visual Features".IEEE TRANSACTIONS ON IMAGE PROCESSING 22.3(2013):860-871.