Importance maps for adaptive information reduction in visual scenes

Research output: Chapter in Book/Report/Conference proceedingConference contribution

9 Citations (Scopus)

Abstract

The concept of an "importance map" is defined, providing a means for modelling the relative importance to the viewer of parts of a visual scene. A computational approach for constructing importance maps is described and is used to perform adaptive JPEG compression of two test images. Compared with conventional JPEG results, the information content is reduced while maintaining acceptable perceived quality for viewers.

Original languageEnglish
Title of host publicationANZIIS 1995 - Proceedings of the 3rd Australian and New Zealand Conference on Intelligent Information Systems
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages24-29
Number of pages6
ISBN (Electronic)0864224303, 9780864224309
DOIs
Publication statusPublished - Dec 1995
Externally publishedYes
Event3rd Australian and New Zealand Conference on Intelligent Information Systems, ANZIIS 1995 - Perth, Australia
Duration: 27 Nov 1995 → …

Publication series

NameANZIIS 1995 - Proceedings of the 3rd Australian and New Zealand Conference on Intelligent Information Systems

Conference

Conference3rd Australian and New Zealand Conference on Intelligent Information Systems, ANZIIS 1995
CountryAustralia
CityPerth
Period27/11/95 → …

Fingerprint Dive into the research topics of 'Importance maps for adaptive information reduction in visual scenes'. Together they form a unique fingerprint.

Cite this