A nonlocal feature-driven exemplar-based approach for image inpainting
This work addresses image inpainting for computer vision applications, presenting an incremental improvement by unifying structure and texture recovery in a single framework.
The authors tackled the problem of simultaneously restoring missing structures and textures in damaged images by developing a nonlocal variational framework that combines convolution operators for geometric structures with exemplar-based texture inpainting. They introduced an anisotropic patch distance metric and an optimization algorithm, demonstrating experimental validity on various test images.
We present a nonlocal variational image completion technique which admits simultaneous inpainting of multiple structures and textures in a unified framework. The recovery of geometric structures is achieved by using general convolution operators as a measure of behavior within an image. These are combined with a nonlocal exemplar-based approach to exploit the self-similarity of an image in the selected feature domains and to ensure the inpainting of textures. We also introduce an anisotropic patch distance metric to allow for better control of the feature selection within an image and present a nonlocal energy functional based on this metric. Finally, we derive an optimization algorithm for the proposed variational model and examine its validity experimentally with various test images.