| 摘要 |
A method of representing motion of an object appearing in a sequence of images comprises deriving for each image a set of representative points representing the location of the object, deriving an approximate function representing the trajectory of a representative point in two or more of said sequence of images, and calculating an error value for said approximate function for the representative point for an image, characterised in that the error value is based on the difference between the area of the object as represented by the representative point and the area of the object with the representative point replaced by the respective approximate function value. <IMAGE> |
