Image inpainting based on low-rank ALM algorithm

Image inpainting is the process of reconstructing lost or deteriorated part of images based on priori knowledge or background information. In this paper, we use a low-rank structure with global information to fix the image, the Augmented Lagrange Multiplier(ALM) to optimize our algorithm. We use the Alternating Direction Method of Multipliers(ADMM) to solve image inpainting problems. According to our experiments, information of image scenes is dominated by the largest singular value of the previous small part. Inspired by this, we use the characterize top 20 largest singular values of the information to repair the image. Our experimental analyses show that, using our method to repair images is better than previous traditional method.

[1]  Pascal Getreuer,et al.  Total Variation Inpainting using Split Bregman , 2012, Image Process. Line.

[2]  Inderjit S. Dhillon,et al.  Guaranteed Rank Minimization via Singular Value Projection , 2009, NIPS.

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

[4]  Yi-Qing Wang,et al.  E-PLE: an Algorithm for Image Inpainting , 2013, Image Process. Line.

[5]  Pablo A. Parrilo,et al.  Guaranteed Minimum-Rank Solutions of Linear Matrix Equations via Nuclear Norm Minimization , 2007, SIAM Rev..

[6]  Yiguang Liu,et al.  A Random Algorithm for Low-Rank Decomposition of Large-Scale Matrices With Missing Entries , 2014, IEEE Transactions on Image Processing.

[7]  G. Sapiro,et al.  A collaborative framework for 3D alignment and classification of heterogeneous subvolumes in cryo-electron tomography. , 2013, Journal of structural biology.

[8]  Lizhi Cheng,et al.  Image inpainting based on low-rank and joint-sparse matrix recovery , 2013 .

[9]  Shanmuganathan Raman,et al.  Robust PCA-based solution to image composition using augmented Lagrange multiplier (ALM) , 2016, The Visual Computer.

[10]  Patrick Pérez,et al.  Non-Local Patch-Based Image Inpainting , 2017, Image Process. Line.

[11]  E. Candès,et al.  Exact low-rank matrix completion via convex optimization , 2008, 2008 46th Annual Allerton Conference on Communication, Control, and Computing.

[12]  Anders P. Eriksson,et al.  Efficient computation of robust low-rank matrix approximations in the presence of missing data using the L1 norm , 2010, 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[13]  In-So Kweon,et al.  Partial Sum Minimization of Singular Values in Robust PCA: Algorithm and Applications , 2015, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[14]  Tae-Hyun Oh.  Fast Randomized Singular Value Thresholding for Nuclear Norm Minimization , 2015 .

[15]  Junfeng Yang,et al.  Linearized augmented Lagrangian and alternating direction methods for nuclear norm minimization , 2012, Math. Comput..

[16]  Y. Zhang,et al.  Augmented Lagrangian alternating direction method for matrix separation based on low-rank factorization , 2014, Optim. Methods Softw..

[17]  Shuicheng Yan,et al.  Practical low-rank matrix approximation under robust L1-norm , 2012, 2012 IEEE Conference on Computer Vision and Pattern Recognition.

[18]  Xuelong Li,et al.  Fast and Accurate Matrix Completion via Truncated Nuclear Norm Regularization , 2013, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[19]  Yi Ma,et al.  The Augmented Lagrange Multiplier Method for Exact Recovery of Corrupted Low-Rank Matrices , 2010, Journal of structural biology.