Region Based Patch Propagation and Patch Inpainting for Image Completion

Film and photography archives nowadays go through an accelerated process of degradation. Since the preservation of cultural heritage plays an important role in our society, photograph/film restoration has drawn a lot of attention recently. In this paper, an extent of exemplar based inpainting at determining patch priority and patch matching is proposed. Patch priority is defined by two terms: the number of homogeneous regions within the patch (heterogeneous term) obtained by image segmentation and the variance of the pixels within each homogeneous region. The first term prioritizes the heterogeneous patches and the second term defines the inconsistency within each homogeneous region of the heterogeneous patch to discriminate patches having similar heterogeneous term. Patch inpainting is done in two stages. First, the missing pixels of the patch under consideration are interpolated and then a weighted patch matching criterion is applied for finding the candidate patch. Experimental results on damaged digitized photographs and natural images are presented, which demonstrate the effectiveness of the image-completion framework for tasks such as scratch/text, object removal and image inpainting.

[1]  Guillermo Sapiro,et al.  Structure and texture filling-in of missing image blocks in wireless transmission and compression applications , 2003, IEEE Trans. Image Process..

[2]  Eero P. Simoncelli,et al.  A Parametric Texture Model Based on Joint Statistics of Complex Wavelet Coefficients , 2000, International Journal of Computer Vision.

[3]  Jiaya Jia,et al.  Image completion with structure propagation , 2005, SIGGRAPH 2005.

[4]  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.

[5]  David Zhang,et al.  Image information restoration based on long-range correlation , 2002, IEEE Trans. Circuits Syst. Video Technol..

[6]  Daniel Cohen-Or,et al.  Fragment-based image completion , 2003, ACM Trans. Graph..

[7]  David Tschumperlé,et al.  Fast Anisotropic Smoothing of Multi-Valued Images using Curvature-Preserving PDE's , 2006, International Journal of Computer Vision.

[8]  Onur G. Guleryuz,et al.  Nonlinear approximation based image recovery using adaptive sparse reconstructions and iterated denoising-part II: adaptive algorithms , 2006, IEEE Transactions on Image Processing.

[9]  Patrick Pérez,et al.  Poisson image editing , 2003, ACM Trans. Graph..

[10]  Guillermo Sapiro,et al.  Simultaneous structure and texture image inpainting , 2003, IEEE Trans. Image Process..

[11]  Charlie C. L. Wang,et al.  Gradient based image completion by solving the Poisson equation , 2007, Comput. Graph..

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

[13]  Nikos Komodakis,et al.  Image Completion Using Efficient Belief Propagation Via Priority Scheduling and Dynamic Pruning , 2007, IEEE Transactions on Image Processing.

[14]  Daniel P. Huttenlocher,et al.  Efficient Graph-Based Image Segmentation , 2004, International Journal of Computer Vision.

[15]  Alexander Wong,et al.  A nonlocal-means approach to exemplar-based inpainting , 2008, 2008 15th IEEE International Conference on Image Processing.

[16]  Zongben Xu,et al.  Image Inpainting by Patch Propagation Using Patch Sparsity , 2010, IEEE Transactions on Image Processing.

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