Image-Word Mapping

View Sample PDF

Author(s): Yang Cai (Carnegie Mellon University, USA)and David Kaufer (Carnegie Mellon University, USA)
Copyright: 2011
Pages: 10
Source title: Handbook of Research on Ambient Intelligence and Smart Environments: Trends and Perspectives
Source Author(s)/Editor(s): Nak-Young Chong (Japan Advanced Institute of Science and Technology, Japan)and Fulvio Mastrogiovanni (University of Genova, Italy)
DOI: 10.4018/978-1-61692-857-5.ch005

Keywords: Artificial Intelligence / Computer Science & IT / Information Science Reference

Purchase

View Image-Word Mapping on the publisher's website for pricing and purchasing information.

Abstract

No Ambient Intelligence can survive without human-computer interactions. Over ninety percent of information in our communication is verbal and visual. The mapping between one-dimensional words and two-dimensional images is a challenge for visual information classification and reconstruction. In this Chapter, we present a model for the image-word two-way mapping process. The model applies specifically to facial identification and facial reconstruction. It accommodates through semantic differential descriptions, analogical and graph-based visual abstraction that allows humans and computers to categorize objects and to provide verbal annotations to the shapes that comprise faces. An image-word mapping interface is designed for efficient facial recognition in massive visual datasets. We demonstrate how a two-way mapping of words and facial shapes is feasible in facial information retrieval and reconstruction.

The IRMA Community

Research IRM

Image-Word Mapping

Purchase

Abstract

Related Content

IRMA Sponsors