Abstract:

Seamless cloning of a source image patch into a target image is an important and useful image editing operation, which has received considerable research attention in recent years. This operation is typically carried out by solving a Poisson equation with different boundary conditions, which smoothly interpolates the discrepancies between the boundary of the source patch and the target across the entire cloned area. In this paper, different approaches like alpha matting, gradient domain, mean value coordinates are introduced. In gradient domain methods obtain the final cloning result in the cloned region by solving a Poisson equation, which eventually solves a large sparse linear system. The results of gradient domain methods are seamless and match well with the destination image on the overall appearance. In coordinate-based approach, where rather than solving a large linear system to perform the aforementioned interpolation, the value of the interpolant at each interior pixel is given by a weighted combination of values along the boundary. Style-aware image cloning allows users to seamlessly insert any photorealistic or artificial objects into an artwork to create a new image that shares the same artistic style with the original artwork. Extensive experimental results demonstrate the effectiveness of these methods.


Keywords — Image editing, image gradient, mean value coordinates, Poisson equation, seamless cloning.