| 摘要 |
A method of aligning a representation of a garment contained in a first image with a representation of a person contained in a second image. The method including allocating nodes to predetermined key points on the garment, analysing the shape of the person to find the outline of the person, allocating nodes to predetermined points on the person, such as the wrists, elbows , shoulders and chest. The first and second images are manipulated to align the predetermined points and the first image is scaled and overlaid onto the second image. A method for analysing an image to locate and isolate an object within the image from a background includes classifying the colour of each pixel of the image, estimating a colour descriptor of the object and the background to isolate the object in the image and identify a shape descriptor of the object. A method of searching for an image containing an object includes determining the complexity of the image, the bounds, shape descriptor and colour descriptor of the object, returning images based on comparison of colour and shape descriptors. |
