Identification based on facial parts

Author:

Stevanov Zorica1,Zdravkovic Suncica1

Affiliation:

1. Laboratorija za eksperimentalnu psihologiju, Filozofski fakultet, Beograd

Abstract

Two opposing views dominate face identification literature, one suggesting that the face is processed as a whole and another suggesting analysis based on parts. Our research tried to establish which of these two is the dominant strategy and our results fell in the direction of analysis based on parts. The faces were covered with a mask and the participants were uncovering different parts, one at the time, in an attempt to identify a person. Already at the level of a single facial feature, such as mouth or eye and top of the nose, some observers were capable to establish the identity of a familiar face. Identification is exceptionally successful when a small assembly of facial parts is visible, such as eye, eyebrow and the top of the nose. Some facial parts are not very informative on their own but do enhance recognition when given as a part of such an assembly. Novel finding here is importance of the top of the nose for the face identification. Additionally observers have a preference toward the left side of the face. Typically subjects view the elements in the following order: left eye, left eyebrow, right eye, lips, region between the eyes, right eyebrow, region between the eyebrows, left check, right cheek. When observers are not in a position to see eyes, eyebrows or top of the nose, they go for lips first and then region between the eyebrows, region between the eyes, left check, right cheek and finally chin.

Publisher

National Library of Serbia

Subject

General Psychology

Cited by 1 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

1. Deidentification of Facial Images Using Composites;Journal of Oral and Maxillofacial Surgery;2011-12

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3