Fast algorithm for color texture image inpainting using the non-local CTV model

The classical non-local Total Variation model has been extensively used for gray texture image inpainting previously, but such model can not be directly applied to color texture image inpainting due to coupling of different image channels in color images. In order to solve the inpainting problem for color texture images effectively, we propose a non-local Color Total Variation model. This model is different from the recently proposed non-local Mumford-Shah model (NL-MS). Technically, the proposed model is an extension of local TV model for gray images but we take account of the relationship between different channels in color images and make use of concepts of the non-local operators. We will analyze how the coupling of different channels of color images in the proposed model makes the problem difficult for numerical implementation with the conventional split Bregman algorithm. In order to solve the proposed model efficiently, we propose a fast heuristic numerical algorithm based on the split Bregman algorithm with introduction of a threshold function. The performance of the proposed model with the proposed heuristic algorithm is compared with the NL-MS model. Extensive numerical experiments have shown that the proposed model and algorithm have superior excellent performance as well as with much faster speed.