Abstract: A system and method is presented for identifying a plurality of products within a collection of product images in an online shopping environment. The method includes providing the collection of images representing products offered for sale; receiving an input search criteria from a prospective customer. The criteria include at least one of a query color. The method also includes identifying and extracting colors within the collection of images based on a presence of significant colors in an image; and comparing and matching the query color to the identified and extracted significant colors. The method continues by determining, within a predetermined range of accuracy, products from within the collection of images having the matched query color; and presenting the determined products to the prospective customer. In one embodiment, the presentation is provided by means for a graphical user interface.

Abstract: A method for generating representations of visual characteristics of images is presented. The method includes receiving search criteria. The criteria include images to be searched, query images and expected result sets, and a retrieval metric. The method identifies objects within each image and selectively generates a representation of visual characteristics of each image using descriptors from an inventory of descriptors in accordance with the retrieval metric. The method compares the representations of the query image to representations of the images to be searched and determines a search result. The search result is compared to the expected result. If the results do not match, the generating, comparing and determining steps are re-executed with reselected descriptors based on the search result and the retrieval metric. The re-execution continues in a trial-and-error approach until acceptable search results are achieved. When achieved, the method encodes the process for generating the representations.