Abstract
In this paper we describe a system that embarks upon the problem of efficient video processing and representation for automatic semantic classification and modelling for indexing and retrieval of large multimedia databases. The major focus of the system is the integration of a large-scale wildlife digital video archive with manually annotated semantic metadata organised in a structured taxonomy and media classification system. Our research challenges the problems of temporal analysis of video, intelligent key-frame extraction and summarisation, animal gait analysis, unsupervised semantic modelling and classification and identification of individual animals. In order to evaluate the system a demonstrator is being developed as a part of the ICBR project within the 3C Research programme of convergent technology research for digital media processing and communications.