From Visual Input to Verbal Output in the Visual Translator

摘要: The project VITRA (VIsual TRAnslator) deals with the relationship between natural language and vision. Experimental studies are being carried out in way of designing an interface image understanding systems, aim developing systems for description sequences. During last ten years several domains application have been investigated approach integrating vision has evolved, which allows incremental evaluation simultaneous real world This contribution provides a compact conception presents some results obtained so far. Current limitations future research tasks will be discussed as well. In: Proc. AAAI Fall Symposium on Computational Models Integrating Language Vision, Cambridge, MA, 1995.