Research on Algorithm for Fast Image Inpainting Based on Exemplars
Shengfeng Zhu · Proceedings of the 2018 3rd International Workshop on Materials Engineering and Computer Sciences (IWMECS 2018) · 2018
An exemplars-based two-step image inpainting algorithm is presented.Firstly, the input image is inpainted synchronically with its coarse version via the patch-GST-based inpainting priority and best-match patches based on the weighted SSD.This inpainted image is plausible except some artifacts in the areas that contain edges.Then, the second-time inpainting proceeds employed only on these areas.In the searching of the best-match patches, patch structural distance and dynamic search window techniques are proposed to improve the searching speed.Experimental results show that the inpainted images obtained by the proposed algorithm can effectively recover the structural continuity of the input image.In addition, the processing speed of the proposed algorithm is much faster than the existing competing algorithms.