Image Inpainting Through Metric Labeling via Guided Patch Mixing

Abstract

In this paper, we present a novel formulation of exemplar-based image inpainting as a metric labeling problem, and solve it through the simulated annealing algorithm. Due to their greedy nature, exemplar-based methods sometimes produce inpainted images, which are visually inconsistent. These methods are highly dependent upon the initialization. To solve these problems, we generate five images with a different initialization. A suitable mixture of these five images produces a good inpainted image. The cost function of the proposed metric labeling problem consists of three components, namely, neighbor cost, total variation cost, and structure cost. A linear combination among these components is used to maintain better visual consistency in the inpainted region having smooth transition from the bordering regions of the source image. We use a quality measure to this end. Our experiments on a wide variety of images demonstrate that the proposed technique produces better inpainting images as compared with some other state-of-the-art techniques.

Inpainting of Structural Reconstruction of Monuments Using Singular Value Decomposition Refinement of Patches

“…Reconstruction of the matrix from the above will give us the values that are refined and selected, the equation for reconstructing the data is given by (11)…”

“…A energy function is added to a data term which is minimized to improve the accuracy of the algorithm. A novel approach for image inpainting is proposed in [11] to solve the inpainting problem using a metric label. A simulated annealing algorithm is used to solve this metric labeling problem to generate images with better visual quality.…”

An Adaptive Face Image Inpainting Algorithm Based on Feature Symmetry

“…For the defective area in the image, starting from the edge of the target area, using the structure of the non-target area and texture information, the unknown area is predicted and patched according to the matching criteria, so that the filled image is visually reasonable and real [6]. According to different principles, digital image inpainting algorithms can be divided into two categories: structural propagation methods based on partial differential equations (PDEs) [7] and texture synthesis methods based on sample block [8].…”

“…The center R i (X i , Y i ) of each set S 1 , S 2 , S 3 , S 4 is calculated. The Euclidean distance between the points in S i and R i is calculated, and the maximum radius r i of the circular domain is determined using Equation (8), where N i represents the number of feature points in S i . Taking R i as the center of the circle and r i as the maximum radius, a circle E i with a maximum radius of the characteristic part is obtained.…”

“…The confidence of the pixels in the repaired area is mainly updated, and then the next pixel is prepared for inpainting. Steps (2)(3)(4)(5)(6)(7)(8) are repeated until the face image is repaired. 9.…”