A new gradient-driven inpainting approach

摘要

To the large-scale region geometric inpainting of the digital image, the gradient-driven computing model is superior to previous other models, but its performance is not stable to different inpainted objects because the information propagating direction is uncertain in the inpainting process. In this paper, a novel conclusion is firstly described that it is decisive to the inpainting result that the information propagating direction always points to the outside of inpainted region. Finally, a new method of gradient-driven image inpainting is proposed. Experimental results prove that the method can stabilize the information propagating direction of the inpainting process and its inpainting performance is more robust.