File(s) under permanent embargo
Multiple classivî¬®vÖer object detection with convî¬®vÖdence measures.
chapterposted on 2023-05-28, 00:47 authored by Horton, M, Cameron-Jones, RM, Williams, RN
This paper describes an extension to the Haar Classivî¬®vÖer Cascade technique for object detection. Existing Haar Classivî¬®vÖer Cascades are binary; the extension adds convî¬®vÖdence measurement. This convî¬®vÖdence measure was implemented and found to improve accuracy on two object detection problems: face detection and vî¬®vÖsh detection. For vî¬®vÖsh detection, the problem of selecting positive training-sample angle-ranges was also considered; results showed that large random variations that result in cascades covering overlapping ranges increases their accuracy.
Publication titleAI 2007: Advances in Artificial Intelligence. Proceedings of the 20th Australian Joint Conference on Artificial Intelligence, Gold Coast, Australia, December 2-6, 2007
SeriesLecture Notes in Computer Science
Place of publicationHeidelberg
Event title20th Australasian Joint Conference on Artificial Intelligence
Event VenueHobart, Tasmania
Date of Event (Start Date)2007-12-02
Date of Event (End Date)2007-12-06
Rights statementThe original publication is available at www.springerlink.com