Mapping between image regions and caption concepts of captioned depictive photographs
Loading...
Authors
Rowe, Neil C.
Subjects
Advisors
Date of Issue
1998-07
Date
July 1998
Publisher
Monterey, California. Naval Postgraduate School
Language
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.
Type
Conference Paper
Description
This paper appeared in the AAAI-98 Workshop on Representations for Multi-Modal Human-Computer
Interaction, July 1998, Madison, Wisconsin, USA.
Series/Report No
Department
Organization
Identifiers
NPS Report Number
Sponsors
supported by the U.S. Army Artificial Intelligence Center, and by the U. S. Naval Postgraduate School
Funder
funds provided by the Chief for Naval Operations
Format
Citation
Distribution Statement
Approved for public release; distribution is unlimited.