Contrast sensitivity functions in autoencoders

Three decades ago, Atick et al. suggested that human frequency sensitivity may emerge from the enhancement required for a more efficient analysis of retinal images. Here we reassess the relevance of low-level vision tasks in the explanation of the contrast sensitivity functions (CSFs) in light of 1)...

Descripción completa

Detalles Bibliográficos
Autores: Li, Qiang, Gomez-Villa, Alex, Bertalmío, Marcelo, Malo, Jesús
Tipo de recurso: artículo
Estado:Versión publicada
Fecha de publicación:2022
País:España
Institución:Varias* (Consorci de Biblioteques Universitáries de Catalunya, Centre de Serveis Científics i Acadèmics de Catalunya)
Repositorio:Recercat. Dipósit de la Recerca de Catalunya
OAI Identifier:oai:recercat.cat:10230/55982
Acceso en línea:http://hdl.handle.net/10230/55982
http://dx.doi.org/10.1167/jov.22.6.8
Access Level:acceso abierto
Palabra clave:spatiotemporal and chromatic contrast sensitivity
convolutional autoencoders
modulation transfer function
noisy cones
deblurring and denoising
chromatic adaptation
natural images
statistical goals
architectures
id ES_64f2f796bfe304349b0dcfa2d90ffd39
oai_identifier_str oai:recercat.cat:10230/55982
network_acronym_str ES
network_name_str España
repository_id_str
spelling Contrast sensitivity functions in autoencodersLi, QiangGomez-Villa, AlexBertalmío, MarceloMalo, Jesússpatiotemporal and chromatic contrast sensitivityconvolutional autoencodersmodulation transfer functionnoisy conesdeblurring and denoisingchromatic adaptationnatural imagesstatistical goalsarchitecturesThree decades ago, Atick et al. suggested that human frequency sensitivity may emerge from the enhancement required for a more efficient analysis of retinal images. Here we reassess the relevance of low-level vision tasks in the explanation of the contrast sensitivity functions (CSFs) in light of 1) the current trend of using artificial neural networks for studying vision, and 2) the current knowledge of retinal image representations. As a first contribution, we show that a very popular type of convolutional neural networks (CNNs), called autoencoders, may develop human-like CSFs in the spatiotemporal and chromatic dimensions when trained to perform some basic low-level vision tasks (like retinal noise and optical blur removal), but not others (like chromatic) adaptation or pure reconstruction after simple bottlenecks). As an illustrative example, the best CNN (in the considered set of simple architectures for enhancement of the retinal signal) reproduces the CSFs with a root mean square error of 11% of the maximum sensitivity. As a second contribution, we provide experimental evidence of the fact that, for some functional goals (at low abstraction level), deeper CNNs that are better in reaching the quantitative goal are actually worse in replicating human-like phenomena (such as the CSFs). This low-level result (for the explored networks) is not necessarily in contradiction with other works that report advantages of deeper nets in modeling higher level vision goals. However, in line with a growing body of literature, our results suggests another word of caution about CNNs in vision science because the use of simplified units or unrealistic architectures in goal optimization may be a limitation for the modeling and understanding of human vision.Partially funded by these grants from GVA/AEI/FEDER/EU: MICINN DPI2017- 89867-C2-2-R, MICINN PID2020-118071GB-I00, and GVA Grisolía-P/2019/035 (for JM and QL), and MICINN PGC2018-099651-B-I00 (for A.G.V. and M.B.).Association for Research in Vision and Ophthalmology (ARVO)202320232022info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionapplication/pdfapplication/pdfhttp://hdl.handle.net/10230/55982http://dx.doi.org/10.1167/jov.22.6.8reponame:Recercat. Dipósit de la Recerca de Catalunyainstname:Varias* (Consorci de Biblioteques Universitáries de Catalunya, Centre de Serveis Científics i Acadèmics de Catalunya)InglésJournal of Vision. 2022;22(6):8.info:eu-repo/grantAgreement/ES/2PE/PID2020-118071GB-I00info:eu-repo/grantAgreement/ES/2PE/DPI2017-89867-C2-2-Rinfo:eu-repo/grantAgreement/ES/2PE/PGC2018-099651-B-I00Copyright 2022 The Authors. This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.http://creativecommons.org/licenses/by-nc-nd/4.0/info:eu-repo/semantics/openAccessoai:recercat.cat:10230/559822026-05-29T05:05:01Z
dc.title.none.fl_str_mv Contrast sensitivity functions in autoencoders
title Contrast sensitivity functions in autoencoders
spellingShingle Contrast sensitivity functions in autoencoders
Li, Qiang
spatiotemporal and chromatic contrast sensitivity
convolutional autoencoders
modulation transfer function
noisy cones
deblurring and denoising
chromatic adaptation
natural images
statistical goals
architectures
title_short Contrast sensitivity functions in autoencoders
title_full Contrast sensitivity functions in autoencoders
title_fullStr Contrast sensitivity functions in autoencoders
title_full_unstemmed Contrast sensitivity functions in autoencoders
title_sort Contrast sensitivity functions in autoencoders
dc.creator.none.fl_str_mv Li, Qiang
Gomez-Villa, Alex
Bertalmío, Marcelo
Malo, Jesús
author Li, Qiang
author_facet Li, Qiang
Gomez-Villa, Alex
Bertalmío, Marcelo
Malo, Jesús
author_role author
author2 Gomez-Villa, Alex
Bertalmío, Marcelo
Malo, Jesús
author2_role author
author
author
dc.subject.none.fl_str_mv spatiotemporal and chromatic contrast sensitivity
convolutional autoencoders
modulation transfer function
noisy cones
deblurring and denoising
chromatic adaptation
natural images
statistical goals
architectures
topic spatiotemporal and chromatic contrast sensitivity
convolutional autoencoders
modulation transfer function
noisy cones
deblurring and denoising
chromatic adaptation
natural images
statistical goals
architectures
description Three decades ago, Atick et al. suggested that human frequency sensitivity may emerge from the enhancement required for a more efficient analysis of retinal images. Here we reassess the relevance of low-level vision tasks in the explanation of the contrast sensitivity functions (CSFs) in light of 1) the current trend of using artificial neural networks for studying vision, and 2) the current knowledge of retinal image representations. As a first contribution, we show that a very popular type of convolutional neural networks (CNNs), called autoencoders, may develop human-like CSFs in the spatiotemporal and chromatic dimensions when trained to perform some basic low-level vision tasks (like retinal noise and optical blur removal), but not others (like chromatic) adaptation or pure reconstruction after simple bottlenecks). As an illustrative example, the best CNN (in the considered set of simple architectures for enhancement of the retinal signal) reproduces the CSFs with a root mean square error of 11% of the maximum sensitivity. As a second contribution, we provide experimental evidence of the fact that, for some functional goals (at low abstraction level), deeper CNNs that are better in reaching the quantitative goal are actually worse in replicating human-like phenomena (such as the CSFs). This low-level result (for the explored networks) is not necessarily in contradiction with other works that report advantages of deeper nets in modeling higher level vision goals. However, in line with a growing body of literature, our results suggests another word of caution about CNNs in vision science because the use of simplified units or unrealistic architectures in goal optimization may be a limitation for the modeling and understanding of human vision.
publishDate 2022
dc.date.none.fl_str_mv 2022
2023
2023
dc.type.none.fl_str_mv info:eu-repo/semantics/article
info:eu-repo/semantics/publishedVersion
format article
status_str publishedVersion
dc.identifier.none.fl_str_mv http://hdl.handle.net/10230/55982
http://dx.doi.org/10.1167/jov.22.6.8
url http://hdl.handle.net/10230/55982
http://dx.doi.org/10.1167/jov.22.6.8
dc.language.none.fl_str_mv Inglés
language_invalid_str_mv Inglés
dc.relation.none.fl_str_mv Journal of Vision. 2022;22(6):8.
info:eu-repo/grantAgreement/ES/2PE/PID2020-118071GB-I00
info:eu-repo/grantAgreement/ES/2PE/DPI2017-89867-C2-2-R
info:eu-repo/grantAgreement/ES/2PE/PGC2018-099651-B-I00
dc.rights.none.fl_str_mv http://creativecommons.org/licenses/by-nc-nd/4.0/
info:eu-repo/semantics/openAccess
rights_invalid_str_mv http://creativecommons.org/licenses/by-nc-nd/4.0/
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
application/pdf
dc.publisher.none.fl_str_mv Association for Research in Vision and Ophthalmology (ARVO)
publisher.none.fl_str_mv Association for Research in Vision and Ophthalmology (ARVO)
dc.source.none.fl_str_mv reponame:Recercat. Dipósit de la Recerca de Catalunya
instname:Varias* (Consorci de Biblioteques Universitáries de Catalunya, Centre de Serveis Científics i Acadèmics de Catalunya)
instname_str Varias* (Consorci de Biblioteques Universitáries de Catalunya, Centre de Serveis Científics i Acadèmics de Catalunya)
reponame_str Recercat. Dipósit de la Recerca de Catalunya
collection Recercat. Dipósit de la Recerca de Catalunya
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869409706071556096
score 15,811543

Similares en LA Referencia