View-symmetric representations of faces in human and artificial neural networks.

IF 2 3区 心理学 Q3 BEHAVIORAL SCIENCES
Neuropsychologia Pub Date : 2025-01-29 Epub Date: 2024-12-05 DOI:10.1016/j.neuropsychologia.2024.109061
Xun Zhu, David M Watson, Daniel Rogers, Timothy J Andrews
{"title":"View-symmetric representations of faces in human and artificial neural networks.","authors":"Xun Zhu, David M Watson, Daniel Rogers, Timothy J Andrews","doi":"10.1016/j.neuropsychologia.2024.109061","DOIUrl":null,"url":null,"abstract":"<p><p>View symmetry has been suggested to be an important intermediate representation between view-specific and view-invariant representations of faces in the human brain. Here, we compared view-symmetry in humans and a deep convolutional neural network (DCNN) trained to recognise faces. First, we compared the output of the DCNN to head rotations in yaw (left-right), pitch (up-down) and roll (in-plane rotation). For yaw, an initial view-specific representation was evident in the convolutional layers, but a view-symmetric representation emerged in the fully-connected layers. Consistent with a role in the recognition of faces, we found that view-symmetric responses to yaw were greater for same identity compared to different identity faces. In contrast, we did not find a similar transition from view-specific to view-symmetric representations in the DCNN for either pitch or roll. These findings suggest that view-symmetry emerges when opposite rotations of the head lead to mirror images. Next, we compared the view-symmetric patterns of response to yaw in the DCNN with corresponding behavioural and neural responses in humans. We found that responses in the fully-connected layers of the DCNN correlated with judgements of perceptual similarity and with the responses of higher visual regions. These findings suggest that view-symmetric representations may be computationally efficient way to represent faces in humans and artificial neural networks for the recognition of identity.</p>","PeriodicalId":19279,"journal":{"name":"Neuropsychologia","volume":" ","pages":"109061"},"PeriodicalIF":2.0000,"publicationDate":"2025-01-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Neuropsychologia","FirstCategoryId":"102","ListUrlMain":"https://doi.org/10.1016/j.neuropsychologia.2024.109061","RegionNum":3,"RegionCategory":"心理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2024/12/5 0:00:00","PubModel":"Epub","JCR":"Q3","JCRName":"BEHAVIORAL SCIENCES","Score":null,"Total":0}
引用次数: 0

Abstract

View symmetry has been suggested to be an important intermediate representation between view-specific and view-invariant representations of faces in the human brain. Here, we compared view-symmetry in humans and a deep convolutional neural network (DCNN) trained to recognise faces. First, we compared the output of the DCNN to head rotations in yaw (left-right), pitch (up-down) and roll (in-plane rotation). For yaw, an initial view-specific representation was evident in the convolutional layers, but a view-symmetric representation emerged in the fully-connected layers. Consistent with a role in the recognition of faces, we found that view-symmetric responses to yaw were greater for same identity compared to different identity faces. In contrast, we did not find a similar transition from view-specific to view-symmetric representations in the DCNN for either pitch or roll. These findings suggest that view-symmetry emerges when opposite rotations of the head lead to mirror images. Next, we compared the view-symmetric patterns of response to yaw in the DCNN with corresponding behavioural and neural responses in humans. We found that responses in the fully-connected layers of the DCNN correlated with judgements of perceptual similarity and with the responses of higher visual regions. These findings suggest that view-symmetric representations may be computationally efficient way to represent faces in humans and artificial neural networks for the recognition of identity.

人类和人工神经网络中人脸的视图对称表示。
视图对称被认为是人脸在人脑中特定视图和不变视图之间重要的中间表征。在这里,我们比较了人类的视觉对称性和经过训练识别人脸的深度卷积神经网络(DCNN)。首先,我们将DCNN的输出与偏航(左右)、俯仰(上下)和滚转(平面内旋转)中的头部旋转进行了比较。对于偏航,最初的特定于视图的表示在卷积层中很明显,但在完全连接的层中出现了视图对称表示。与面孔识别中的作用一致,我们发现,与不同身份的面孔相比,相同身份的人对偏航的视图对称反应更大。相比之下,我们在俯仰或翻滚的DCNN中没有发现类似的从视图特定表示到视图对称表示的转变。这些发现表明,当头部的相反旋转导致镜像时,视图对称性就会出现。接下来,我们比较了DCNN对偏航的视觉对称反应模式与人类相应的行为和神经反应。我们发现,DCNN全连接层的反应与感知相似性的判断和更高视觉区域的反应相关。这些发现表明,视图对称表示可能是一种计算效率很高的方法来表示人类和人工神经网络的面孔,以识别身份。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
Neuropsychologia
Neuropsychologia 医学-行为科学
CiteScore
5.10
自引率
3.80%
发文量
228
审稿时长
4 months
期刊介绍: Neuropsychologia is an international interdisciplinary journal devoted to experimental and theoretical contributions that advance understanding of human cognition and behavior from a neuroscience perspective. The journal will consider for publication studies that link brain function with cognitive processes, including attention and awareness, action and motor control, executive functions and cognitive control, memory, language, and emotion and social cognition.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信