Interactive image deformation using cage coordinates on GPU
Weiliang Meng, Bin Sheng, Shandong Wang, Hanqiu Sun, Enhua Wu · 2009
Cage-based deformation is widely used in various applications of image and geometry processing. In this paper, we show how to deform a 2D image by using Cage Coordinates (CC), which naturally unifies the existing coordinates, such as Mean Value Coordinates (MVC), Harmonic Coordinates (HC) and Green Coordinates (GC) into the same formulation. The cage is generated interactively for the region-of-interest (ROI), and the warping effect is obtained by simply changing the positions of the cage vertices. An intuitive method is designed to keep the image shape during the process of deformation. Meanwhile, a new GPU-based framework of using CC for real time image deformation is developed. Such real-time rendering and the intuitive user interaction in our cage-based image deformation make it easy to conduct the image animation by enabling the cage vertices' 2D motions, as demonstrated in the experimental results of the proposed technique.