Vlad Morariu is an assistant research scientist and a member of the Center for Automation Research.
His research focuses on high-level common sense reasoning about the visual world in the face of observation and model uncertainty.
Morariu's goal is to build a system that can answer complex visual queries that are expressed linguistically and are not known ahead of time. This involves investigating knowledge representations that manage the trade-off between expressiveness and computational complexity; learning rich spatial and temporal models while managing the trade-off between the amount of model structure, training data, and supervision; and reducing uncertainty in mid- and low-level computer vision techniques.
He received his doctorate in computer science from the University of Maryland.
There are no publications listed for this person.