Mapping between image regions and caption concepts of captioned depictive photographs

Rowe, Neil C.

dc.contributor.author	Rowe, Neil C.
dc.date	July 1998
dc.date.accessioned	2013-09-20T16:11:15Z
dc.date.available	2013-09-20T16:11:15Z
dc.date.issued	1998-07
dc.identifier.uri	https://hdl.handle.net/10945/36566
dc.description	This paper appeared in the AAAI-98 Workshop on Representations for Multi-Modal Human-Computer Interaction, July 1998, Madison, Wisconsin, USA.	en_US
dc.description.abstract	We discuss the obstacles to inference of correspondances between objects within photographic images and their counterpart concepts in descriptive captions of those images. This is important for information retrieval of photographic data since its content analysis is much harder than linguistic analysis of its captions. We argue that the key mapping is between certain caption concepts representing the "linguistic focus" and certain image regions representing the "visual focus". The mapping is one-to-many, however, and many image regions and captions concepts are not mapped at all. We discuss some domain-independent constraints that can restrict potential mappings. We also report on experiments testing our criteria for visual focus of images.	en_US
dc.description.sponsorship	supported by the U.S. Army Artificial Intelligence Center, and by the U. S. Naval Postgraduate School	en_US
dc.publisher	Monterey, California. Naval Postgraduate School	en_US
dc.title	Mapping between image regions and caption concepts of captioned depictive photographs	en_US
dc.type	Conference Paper	en_US
dc.description.funder	funds provided by the Chief for Naval Operations	en_US
dc.description.distributionstatement	Approved for public release; distribution is unlimited.