Multi-domain data modeling for biometrics

Recently, much work has been performed on CBIR (content based image retrieval) that treats images as single data domain. However, in our highly digitized society, information is being supplied in multiple domains where the data is linked across domains. For example, a web site does contain images, but it may also contain text, hyperlinks, documents, sound files, movies, and other domains of data. Performing recall operations within single domains eliminates the possibility of employing cross-domain inferences. In this work, a multi-domain search space is presented in with two domains: speech and facial images. A single search space is created that contains data from these vastly different domains and cross-domain inferences are allowed. In other words, queries in the speech domain can retrieve image data even if there was no hard link between these data samples. Generation of multidomain search spaces will eventually expand CBIR systems to include data from a variety of sources.