期刊文献+

最小移动二乘抠图 被引量:3

Moving Least Square Matting
下载PDF
导出
摘要 交互式抠图技术在有限的用户交互下抠取图像的前景,被广泛地应用在图像及视频编辑、三维重建等领域中,有极高的应用价值。近年来的抠图技术中,拉氏矩阵给出alpha图上像素间的线性关系,对alpha图的估计起到了重要作用。提出了一种改进拉普拉斯矩阵的方法,使用移动最小二乘法替代最小二乘法,结合最近邻(KNN)方法给出移动拉氏矩阵,并使用移动拉氏矩阵计算alpha图。实验结果证明了移动拉氏矩阵的有效性。 Interactive alpha matting technique extracts the foreground of image with limited interaction. It is widely applied to image editing,3D construction, and has high applicative value. In recent years, the Matting Laplace matrix gives the linear relation on alpha matte, and takes important roles in alpha matting. In this paper, the Moving Matting Matrix is given, which use moving least square instead of least square method, use KNN( K nearest neighbor) model instead of local neighbor model. We solve the alpha matte by Moving Matting Matrix. Experiments prove this method is effective.
出处 《电视技术》 北大核心 2015年第19期24-27,31,共5页 Video Engineering
基金 浙江省自然科学基金面上项目(LY13F020019) 国家自然科学基金重点项目(61133009) 浙江省大学生科技创新活动计划项目(新苗人才计划)(2013R429004)
关键词 图像抠图 拉氏矩阵 最小移动二乘 image matting matting laplace matrix moving least square
  • 相关文献

参考文献14

  • 1LEVIN A,LISCHINSKI D, WEISS Y. A closed form solution to nat- ural image matting[ C ]//Proc. IEEE Computer Society Conterence on Computer Vision and Pattern Recognition. Los Alamitos: IEEE Press, 2006 : 61-68.
  • 2GASTAL E S L,OI,IVEIRA M M. Shared sampling tbr real-time al- pha matting [ J ]. Computer Graphics Forum, 2010, 29 ( 2 ) : 575 -584.
  • 3SHAHRIAN E, RAJAN D,PRICE B,et. al. Improving image matting using comprehensive sampling sets[C]//Proc. IEEE Conference on Computer Vision and Pattern Recognition. Los Alamitos: IEEE Press, 2013 : 636-643.
  • 4GRADY L,SCHIWIETZ T,AHARON S,et. al. Randomwalks for in- teractive alpha-maning[ C]//Proc. 5th lASTED International Con- fcrence on Visualization, Imaging and hnage Processing. Calgary: ACTA Press,2005:423-429.
  • 5HE K M,SUN J,TANG X O. Fast matting using large kernel matting laplacian matrices[ C ]//Proc. IEEE Conference on Computer Vision and Pattern Recognition. Los Alamitos: IEEE Press. 2010: 2165-2172.
  • 6ZHANG Z P, ZHU Q S, XIE Y Q,l,earning based alpha matting u- sing support vector regression[ C ]//Proc. 19th IEEE Conference on International hnage Processing. Los Alamitos: IEEE Press, 2012: 2109-2112.
  • 7LEE P,WU Y. Nonlocal matting[ C]//Proc. IEEE Conference on Computer Vision and Pattern Recognition. Los Alamitos:IEEE Press, 2011:2193-2200.
  • 8CHEN Q F, LID Z Y, TANG C K. Knn matting[ C]//Proc. IEEE CVPR ( 2012 ). Los Alamitos : IEEE Press ,2012 : 869-876.
  • 9沈皓.基于改进KNN算法的AVS到H.264/AVC快速转码方法[J].电视技术,2015,39(8):35-39. 被引量:1
  • 10CItEN X W, ZHOU D Q, ZHOU Z Y, et. al. Image matting with local and nonlocal smooth priors [ C ]//Pmc. IEEE Conference on Computer Vision and Pattern Recognition. Los Alamitos: IEEE Press, 2013 :1902 - 1907.

二级参考文献64

  • 1杜奇,向健勇,袁胜春.基于边缘强度的红外图像阈值分割方法研究[J].红外与激光工程,2004,33(3):288-291. 被引量:18
  • 2袁基炜,史忠科.一种快速运动目标的背景提取算法[J].计算机应用研究,2004,21(8):128-129. 被引量:15
  • 3陶唐飞,韩崇昭,代雪峰,段战胜.综合边缘检测和区域生长的红外图像分割方法[J].光电工程,2004,31(10):50-52. 被引量:24
  • 4贾云涛,胡事民.基于图切分的交互式图像染色算法[J].计算机学报,2006,29(3):508-512. 被引量:15
  • 5Lieberman H. How to color in a coloring book[CJ // Proceedings of the 5th Annual Conference on Computer Graphics and Interactive Techniques. New York: ACM Press. 1978: 111-116.
  • 6Sykora D. BuridnekJ. ZaraJ. Colorization of black- and-white cartoons[J]. Image and Vision Computing. 2005. 23 (9): 767-782.
  • 7Levin A, Lischinski D, Weiss Y. Colorization using optimization[J]. ACM Transactions on Graphics, 2004. 23 (3): 689-694.
  • 8Yatziv L. Sapiro G. Fast image and video colorization using chrominance blending[J]. IEEE Transactions on Image Processing. 2006. IS(5): 1120-1129.
  • 9Qu Y G. Wong T T. Heng P A. Manga colorization[J]. ACM Transactions on Graphics, 2006. 2S(3): 1214-1220.
  • 10Luan Q, Wen F. Cohen-Or D. et al . Natural image colorization[CJ //Proceedings of Eurographics Symposium on Rendering Techniques. Airc-Ia-Ville , Eurographics Asso-ciation Press, 2007: 309-320.

共引文献10

同被引文献8

引证文献3

二级引证文献5

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部