Illumination invariance and object model in content-based image and video retrieval

With huge amounts of multimedia information connected to the global informa tion network (Internet), efficient and effective image retrieval from large image and video repositories has become an imminent research issue. This a rticle presents our research in the C-BIRD (content-based image retrieval i n digital-libraries) project. In addition to the use of common features suc h as color, texture, shape, and their conjuncts, and the combined content-b ased and description-based techniques, it is shown that (a) color-channel-n ormalization enables search by illumination invariance, and (b) feature loc alization and a three-step matching algorithm (color hypothesis, texture su pport, shape verification) facilitate search by object model in image and v ideo databases. (C) 1999 Academic Press.