Image inpainting via singular value thresholding

This paper proposes an inpainting technique based on the singular value thresholding (SVT) algorithm for matrix completion. The SVT algorithm computes the missing elements of a matrix so that the resultant matrix is an approximate of the completion with least rank. It can be implemented by a straightforward iteration procedure which performs singular value decompositions followed by a soft thresholding on the singular values. Experimental results show that the technique can effectively recover texture parts of damaged images.

[1]  Michael Ashikhmin,et al.  Synthesizing natural textures , 2001, I3D '01.

[2]  Guillermo Sapiro,et al.  Navier-stokes, fluid dynamics, and image and video inpainting , 2001, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.

[3]  Patrick Pérez,et al.  Region filling and object removal by exemplar-based image inpainting , 2004, IEEE Transactions on Image Processing.

[4]  Guillermo Sapiro,et al.  Image inpainting , 2000, SIGGRAPH.

[5]  Jean-Michel Morel,et al.  Level lines based disocclusion , 1998, Proceedings 1998 International Conference on Image Processing. ICIP98 (Cat. No.98CB36269).

[6]  Emmanuel J. Candès,et al.  Exact Matrix Completion via Convex Optimization , 2008, Found. Comput. Math..

[7]  Emmanuel J. Candès,et al.  A Singular Value Thresholding Algorithm for Matrix Completion , 2008, SIAM J. Optim..

[8]  Guillermo Sapiro,et al.  Filling-in by joint interpolation of vector fields and gray levels , 2001, IEEE Trans. Image Process..

[9]  Aurélie Bugeau,et al.  Combining Texture Synthesis and Diffusion for Image Inpainting , 2009, VISAPP.

[10]  Tony F. Chan,et al.  Nontexture Inpainting by Curvature-Driven Diffusions , 2001, J. Vis. Commun. Image Represent..

[11]  Sung Yong Shin,et al.  On pixel-based texture synthesis by non-parametric sampling , 2006, Comput. Graph..

[12]  Guillermo Sapiro,et al.  A Comprehensive Framework for Image Inpainting , 2010, IEEE Transactions on Image Processing.