PROJECT TITLE:

Context-Aware Patch-Based Image Inpainting Using Markov Random Field Modeling - 2015

ABSTRACT:

In this paper, we initial introduce a general approach for context-aware patch-based mostly image inpainting, where textural descriptors are used to guide and accelerate the hunt for well-matching (candidate) patches. A completely unique high-down splitting procedure divides the image into variable size blocks consistent with their context, constraining thereby the rummage around for candidate patches to nonlocal image regions with matching context. This approach can be employed to boost the speed and performance of nearly any (patch-based mostly) inpainting methodology. We have a tendency to apply this approach to the therefore-called world image inpainting with the Markov random field (MRF) previous, where MRF encodes a priori data about consistency of neighboring image patches. We solve the resulting optimization downside with an economical low-complexity inference technique. Experimental results demonstrate the potential of the proposed approach in inpainting applications like scratch, text, and object removal. Improvement and significant acceleration of a connected international MRF-primarily based inpainting method is also evident.