An inpainting method for face images

The digital inpainting process is a process of filling the unknown region of an image and its aim is to inpaint the damaged image and make the observer not aware of that this image has been damaged off and has been inpainted. In this paper, Criminisi algorithm has been studied and improved. In the improved method the gradient term is taken into account together with confidence term and data term in calculating the priority. Thus more information of known pixels is considered in the improved algorithm. Then an inpainting method for face images is proposed based on the improved algorithm. It does not need to spend too much time analyzing the facial features, and can achieve reasonable results.

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

[2]  Rafael C. González,et al.  Local Determination of a Moving Contrast Edge , 1985, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[3]  Patrick Pérez,et al.  Object removal by exemplar-based inpainting , 2003, 2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2003. Proceedings..

[4]  J. Tao,et al.  Reconstruction of Partially Occluded Face by Fast Recursive PCA , 2007, 2007 International Conference on Computational Intelligence and Security Workshops (CISW 2007).

[5]  Sang Min Yoon,et al.  Detection of Partially Occluded Face Using Support Vector Machines , 2002, MVA.

[6]  Kevin Curran,et al.  A skin tone detection algorithm for an adaptive approach to steganography , 2009, Signal Process..

[7]  Allen Y. Yang,et al.  Robust Face Recognition via Sparse Representation , 2009, IEEE Transactions on Pattern Analysis and Machine Intelligence.

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