We present a "parts and structure" model for object category recognition that can be learnt efficiently and in a weakly-supervised manner: the model is learnt from examp...
Annotation-based discussions are an important concept for today's digital libraries and those of the future, containing additional information to and about the content managed...
Scientific data of importance to biologists in the Humitn Genome Project resides not only in conventional da.tabases, but in structured files maintained in a number of different f...
Peter Buneman, Susan B. Davidson, Kyle Hart, G. Ch...
This paper presents an architecture for a persistent object store in which multi-level storage is explicitly included. Traditionally, DBMSs have assumed that all accessible data r...
Abstract. We address the problem of visual event recognition in surveillance where noise and missing observations are serious problems. Common sense domain knowledge is exploited t...