Image morphing deals with the metamorphosis of an image to another image. Using two images, the metamorphosis process produces a sequence of inbetween images in which one image gradually changes into the other over a time interval. Image morphing techniques have been widely used in creating special effects for television commercials, music videos, and movies.
This thesis introduces new approach to image morphing to overcome the problem in previous techniques. When the features are overlapped in an image, positional constraints given to those features can be different from each other in magnitude and direction. In the previous morphing methods which process the image as a whole, the different positional constraints of neighboring features can not be satisfied simultaneously. In order to solve this problem, the proposed method separates overlapped features into new images, and then warp them according to the corresponding positional constraints. Finally the warped feature images are combined to one image. Through the experiment, the method is proven to provide more prospective results compared with the previous methods.
