Face Space Representations in Deep Convolutional Neural Networks.

Autor: O'Toole AJ; School of Behavioral and Brain Sciences, The University of Texas at Dallas, Richardson, TX, USA. Electronic address: otoole@utdallas.edu., Castillo CD; University of Maryland Institute for Advanced Computer Studies, College Park, MD, USA., Parde CJ; School of Behavioral and Brain Sciences, The University of Texas at Dallas, Richardson, TX, USA., Hill MQ; School of Behavioral and Brain Sciences, The University of Texas at Dallas, Richardson, TX, USA., Chellappa R; University of Maryland Institute for Advanced Computer Studies, College Park, MD, USA.
Jazyk: angličtina
Zdroj: Trends in cognitive sciences [Trends Cogn Sci] 2018 Sep; Vol. 22 (9), pp. 794-809. Date of Electronic Publication: 2018 Aug 07.
DOI: 10.1016/j.tics.2018.06.006
Abstrakt: Inspired by the primate visual system, deep convolutional neural networks (DCNNs) have made impressive progress on the complex problem of recognizing faces across variations of viewpoint, illumination, expression, and appearance. This generalized face recognition is a hallmark of human recognition for familiar faces. Despite the computational advances, the visual nature of the face code that emerges in DCNNs is poorly understood. We review what is known about these codes, using the long-standing metaphor of a 'face space' to ground them in the broader context of previous-generation face recognition algorithms. We show that DCNN face representations are a fundamentally new class of visual representation that allows for, but does not assure, generalized face recognition.
(Copyright © 2018 Elsevier Ltd. All rights reserved.)
Databáze: MEDLINE
Externí odkaz: