A global Sampling Based Image Matting using Non-negative Matrix Factorization. | [Mehran University Research Journal of Engineering and Technology • 2017]

Author(s):

1. NAVEED ALAM: Department of Computer Science, Federal Urdu University of Arts, Science & Technology, Karachi, Pakistan

2. Muhammad Sarim: Department of Computer Science, Federal Urdu University of Arts, Science & Technology, Karachi, Pakistan

3. Abdul Basit Shaikh: Department of Computer Science, Federal Urdu University of Arts, Science & Technology, Karachi, Pakistan

Abstract:

Image matting is a technique in which a foreground is separated from the background of a given image along with the pixel wise opacity. This foreground can then be seamlessly composited in a different background to obtain a novel scene. This paper presents a global non-parametric sampling algorithm over image patches and utilizes a dimension reduction technique known as NMF (Non-Negative Matrix Factorization). Although some existing non-parametric approaches use large nearby foreground and background regions to sample patches but these approaches fail to take the whole image to sample patches. It is because of the high memory and computational requirements. The use of NMF in the proposed algorithm allows the dimension reduction which reduces the computational cost and memory requirement. The use of NMF also allow the proposed approach to use the whole foreground and background region in the image and reduces the patch complexity and help in efficient patch sampling. The use of patches not only allows the incorporation of the pixel colour but also the local image structure. The use of local structures in the image is important to estimate a high-quality alpha matte especially in the images which have regions containing high texture. The proposed algorithm is evaluated on the standard data set and obtained results are comparable to the state-of-the-art matting techniques.

Page(s): 763-768

DOI: DOI not available

Published: Journal: Mehran University Research Journal of Engineering and Technology, Volume: 36, Issue: 4, Year: 2017

Keywords:

NonParametric SamplingMethod , NonNegative Matrix Factorization

References:

[1] Porter , T.,Duff , T., 1984.,ACM SIGGRAPH 253 -259

[2] Sun , J.,Jia , J.,Tang , C.K., 2004.,ACM Transactions on Graphics 23 315 -321

[3] Grady , L.,Schiwietz , T.,Aharon , S.,Westermann,A. , Lischinski,D. , and Weiss, 2006.“A Closed form Solution to Natural Image Matting”, 61 -68

[4] Ruzon , M.A.,Tomasi , C.,HiltonHead,SC, 2000.Alpha Estimation in Natural Images”, 18 -25

[5] Chuang , Y.Y.,Curless , B.,Salesin , D.H.,Szeliski,CVPR,HI, 2001.A Bayesian Approach to Digital Matting”, 2 264 -271

[6] Hillman , P.,Hannah,Scotland, 2005.,Vision and Graphics 211 -18

[7] Kun , W.,Zheng , N.N.,Liu , W.X., 2005.Natural Image Matting with Nonnegative Matrix Factorization”, 1186 -1189

[8] Wang , J.,M. F., 2007.Optimized Color Sampling for Robust Matting”, 1 -8

[9] Rhemann , C.,Rother , C.,Gelautz , M., 2008.Improving Color Modeling for Alpha Matting”, -

[10] Sarim , M.,Hilton , A.,Guillemaut,Kim,Non-Parametric Natural, 2009.,International Conference on Image Processing ICIP -

[11] Eduardo S.L.G. , and Oliveira,M.M., 2010.Shared Sampling for Real-Time Alpha Matting”,Proceedings of Eurographics Computer Graphics Forum 29 575 -584

[12] Lee , D.D., 1999.Learning the Parts of Objects by Nonnegative Matrix Factorization”,Nature 6755 788 -791

[13] Berman , A.,Vlahos , P.,Dadourian , A., 2000.Comprehensive Method for Removing from an Image the Background Surrounding a Selected Object”,US Patent 134 -

[14] Wang , Y.,No., 2012.Non-Negative Matrix Factorization: AComprehensive Review”,IEEE Transactions on Knowledge and Data Engineering 99 -

[15] Rhemann , C.,Rother , C.,Wang , J.,Gelautz , M.,Kohli , P.,Rott , P., 2009.A Perceptually Motivated Online Benchmark for Image Matting”, 1826 -1833

[16] Levin, A., Lischinski, D., and Weiss, Y., “A Closed form Solution to Natural Image Matting”, CVPR, pp. 61-68, New York, USA, 2006.

Citations

Downloads

Views