METHOD AND APPARATUS FOR RETRIEVING VISUAL OBJECT CATEGORIES FROM A DATABASE CONTAINING IMAGES

摘要

A method for determining the relevance of images retrieved from a database using, for example, a search engine, wherein a visual object category is transformed into a model defining features of a visual object category and a spatial relationship therebetween. Thus "visual consistency" amongst the images that are returned by a search is measured and the images are re-ranked on the basis of this consistency, thereby increasing the proportion of relevant images returned to the user within the first two entries in the search results.