Professor Vasconcelos' research interests are at the intersection of computer vision, and machine learning. The main goal is to provide computers with the ability to perceive complex visual scenes and make inferences from them. Main topics of research include the computational modeling of biological vision systems; object recognition and tracking; action recognition; surveillance of crowded environments; multimedia search, classification, and retrieval; medical imaging; and the design of machine learning algorithms. Possible applications include non-text extensions to current web search engines; devices for personalized access to large video repositories (e.g. movie or newscast databases); advanced surveillance systems; data mining; automated clinical diagnosis systems; etc.