smoothness-constrained face photo-sketch synthesis using sparse representation

ISCAS OpenIR

	smoothness-constrained face photo-sketch synthesis using sparse representation
	Chang Liang; Deng Xiaoming; Zhou Mingquan; Duan Fuqing; Wu Zhongke
	2012
会议名称	21st International Conference on Pattern Recognition, ICPR 2012
会议录名称	Proceedings - International Conference on Pattern Recognition
页码	3025-3029
会议日期	November 11, 2012 - November 15, 2012
会议地点	Tsukuba, Japan
收录类别	EI
ISSN	1051-4651
ISBN	9784990644109
部门归属	(1) College of Information Science and Technology Beijing Normal University China; (2) Institute of Software Chinese Academy of Sciences China
摘要	Face photo-sketch and sketch-photo synthesis have important usages in law enforcement. It is challenging to synthesize face sketches from photos because the drawing techniques and styles of artists' depictions are hard to be learned. To synthesize face photos from sketches is also hard due to its ill-posed nature. In order to avoid mosaic effects in the existed photo-sketch methods, we propose a smoothness-constrained photo-sketch synthesis method via sparse representation. The work is an extension of the previous work[1]. The method is modeled as the minimization of an energy function, a large scale convex optimization problem with l1-norm constraint. Since previous optimization methods are infeasible to solve our problem, we propose an iterative optimization approach, which decomposes the large scale optimization into a sequence of small scale optimizations and solve them iteratively to obtain the approximated optimal solution. The same synthesis strategy can be also used to synthesize photos from sketches. Experiments show its effectiveness. © 2012 ICPR Org Committee.; Face photo-sketch and sketch-photo synthesis have important usages in law enforcement. It is challenging to synthesize face sketches from photos because the drawing techniques and styles of artists' depictions are hard to be learned. To synthesize face photos from sketches is also hard due to its ill-posed nature. In order to avoid mosaic effects in the existed photo-sketch methods, we propose a smoothness-constrained photo-sketch synthesis method via sparse representation. The work is an extension of the previous work[1]. The method is modeled as the minimization of an energy function, a large scale convex optimization problem with l1-norm constraint. Since previous optimization methods are infeasible to solve our problem, we propose an iterative optimization approach, which decomposes the large scale optimization into a sequence of small scale optimizations and solve them iteratively to obtain the approximated optimal solution. The same synthesis strategy can be also used to synthesize photos from sketches. Experiments show its effectiveness. © 2012 ICPR Org Committee.
关键词	Convex Optimization Optimization Pattern Recognition
主办者	Science Council of Japan; Information Processing Society of Japan (IPSJ); Inst. Electron., Inf. Commun. Eng. (IEICE) Inf. Syst. Soc. (ISS); Japan Society for the Promotion of Science (JSPS); The Telecommunications Advancement Foundation
语种	英语
内容类型	会议论文
URI标识	http://ir.iscas.ac.cn/handle/311060/15982
专题	中国科学院软件研究所
推荐引用方式 GB/T 7714	Chang Liang,Deng Xiaoming,Zhou Mingquan,et al. smoothness-constrained face photo-sketch synthesis using sparse representation[C],2012:3025-3029.