Date of Publication :21st June 2017
Abstract: Images need to be retargeted to different sizes and shapes including nonrectangular shapes. This paper proposes a retargeting algorithm that retargets images to a large family of nonrectangular shapes specified by the application. The main objective of this algorithm is preservation of the image content during transformation. This algorithm is based on removing segments of seams based on a cost function. This cost function is calculated by using the image and the target shapes as inputs. This removal reshapes the boundary of the image into nonrectangular shapes. This algorithm can retarget effectively to a large family of nonrectangular shapes known as bhv-convex shapes. This mechanism is efficient in computation and flexible to use
Reference :