Main

Research

My main interest lies in the way we humans see and interpret the world, and how we can create computer systems that are able to mimic these abilities.

I am currently focussing on the linguistic properties of image captions - are these different from "normal" sentences? And if so, can we use this information for automatic image classification?