Reconstructing natural images from human fMRI by alternating encoding and decoding with shared autoencoder regularization

•We proposed the mutual promotion of visual encoding and reconstruction models.•We designed alternating optimization bassed on seim-supervised learning.•We devised inter-sample differentiated representations to augment small dataset.•Our proposed model achieved state-of-the-art reconstruction perfor...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Biomedical signal processing and control Jg. 73; S. 103397
Hauptverfasser: Qiao, Kai, Chen, Jian, Wang, Linyuan, Zhang, Chi, Tong, Li, Yan, Bin
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Elsevier Ltd 01.03.2022
Schlagworte:
ISSN:1746-8094, 1746-8108
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Abstract •We proposed the mutual promotion of visual encoding and reconstruction models.•We designed alternating optimization bassed on seim-supervised learning.•We devised inter-sample differentiated representations to augment small dataset.•Our proposed model achieved state-of-the-art reconstruction performance. Reconstructing the viewed natural images from the corresponding functional magnetic resonance imaging (fMRI) of human visual cortices is extremely difficult. Utilizing deep learning techniques, the quality of reconstructed images can be significantly improved. This however is subject to availability of sufficient number of pair samples, which is not currently the case. In this study, we propose to perform alternating encoding and decoding to realize the mutual promotion of both based on shared semi-supervised learning and accomplish better reconstruction of natural images from the corresponding fMRI voxels. In our proposed method, the encoder and decoder are used to respectively map the images to the fMRI voxels (visual encoding), and the fMRI voxels to the images (visual reconstruction). We argue that combining the encoder and decoder in different sequences can form two converse and shared autoencoders to regularize the supervised learning of both through the unsupervised learning of fMRI voxels and images. More importantly, we alternatingly train the encoder and decoder with shared autoencoder regularization. Here, the premise is that a better encoder can produce a better decoder, and vice versa. The pixel-level identification of the proposed method achieves up to 89.5%, which indicates at least 2% superiority and is the new state-of-the-art compared to the previous works in terms of the image reconstruction performance.
AbstractList •We proposed the mutual promotion of visual encoding and reconstruction models.•We designed alternating optimization bassed on seim-supervised learning.•We devised inter-sample differentiated representations to augment small dataset.•Our proposed model achieved state-of-the-art reconstruction performance. Reconstructing the viewed natural images from the corresponding functional magnetic resonance imaging (fMRI) of human visual cortices is extremely difficult. Utilizing deep learning techniques, the quality of reconstructed images can be significantly improved. This however is subject to availability of sufficient number of pair samples, which is not currently the case. In this study, we propose to perform alternating encoding and decoding to realize the mutual promotion of both based on shared semi-supervised learning and accomplish better reconstruction of natural images from the corresponding fMRI voxels. In our proposed method, the encoder and decoder are used to respectively map the images to the fMRI voxels (visual encoding), and the fMRI voxels to the images (visual reconstruction). We argue that combining the encoder and decoder in different sequences can form two converse and shared autoencoders to regularize the supervised learning of both through the unsupervised learning of fMRI voxels and images. More importantly, we alternatingly train the encoder and decoder with shared autoencoder regularization. Here, the premise is that a better encoder can produce a better decoder, and vice versa. The pixel-level identification of the proposed method achieves up to 89.5%, which indicates at least 2% superiority and is the new state-of-the-art compared to the previous works in terms of the image reconstruction performance.
ArticleNumber 103397
Author Yan, Bin
Zhang, Chi
Chen, Jian
Qiao, Kai
Wang, Linyuan
Tong, Li
Author_xml – sequence: 1
  givenname: Kai
  surname: Qiao
  fullname: Qiao, Kai
  email: qiaokai1992@gmail.com
– sequence: 2
  givenname: Jian
  surname: Chen
  fullname: Chen, Jian
– sequence: 3
  givenname: Linyuan
  surname: Wang
  fullname: Wang, Linyuan
– sequence: 4
  givenname: Chi
  surname: Zhang
  fullname: Zhang, Chi
– sequence: 5
  givenname: Li
  surname: Tong
  fullname: Tong, Li
– sequence: 6
  givenname: Bin
  surname: Yan
  fullname: Yan, Bin
  email: ybspace@hotmail.com
BookMark eNp9kMtKAzEUhoNUsFVfwFVeYGoyk7mBGyleChWh6DpkkpM2ZZopSUapT2-mrRsXXeUk_F84_zdBI9tZQOiOkikltLjfTBu_k9OUpDQ-ZFldXqAxLVmRVJRUo7-Z1OwKTbzfEMKqkrIx2i9BdtYH18tg7ApbEXonWmy2YgUea9dt8brfCov123KOmz0WbQAXY0MarOzUMAirsILT5duENfZr4UBh0YfukAKHHaz6VjjzE-HO3qBLLVoPt6fzGn0-P33MXpPF-8t89rhIZEZISDJdgmB1U2iWS6JYCWWeqZo1DWE6V1LmmmaxtmKpSnOV1SqnWulCpbWQMZ5do_T4r3Sd9w4037nYzu05JXyQxzd8kMcHefwoL0LVP0iacFg7OGHa8-jDEYVY6suA416aqACUcSADV505h_8Cwp-Qbg
CitedBy_id crossref_primary_10_1016_j_jneumeth_2024_110269
crossref_primary_10_3390_brainsci14030234
crossref_primary_10_1016_j_bspc_2024_107110
crossref_primary_10_1109_TNSRE_2023_3283405
Cites_doi 10.1371/journal.pcbi.1006633
10.1109/CVPR42600.2020.00698
10.1007/s11263-015-0816-y
10.1162/NECO_a_00047
10.3390/app9224749
10.1016/j.neucom.2019.12.130
10.1016/j.neuroimage.2018.07.043
10.1016/j.neuron.2008.11.004
10.32470/CCN.2018.1031-0
10.1016/j.cub.2011.08.031
10.1523/JNEUROSCI.5023-14.2015
10.1109/LGRS.2018.2799232
10.1016/j.neuroimage.2010.07.073
10.1109/IJCNN48605.2020.9206960
10.1016/j.neuron.2009.09.006
10.1038/ncomms15037
10.1016/j.neuroimage.2019.05.039
10.1145/3386252
10.3389/fnhum.2018.00242
10.1109/CVPR42600.2020.00674
10.1016/j.tics.2006.07.005
ContentType Journal Article
Copyright 2021 Elsevier Ltd
Copyright_xml – notice: 2021 Elsevier Ltd
DBID AAYXX
CITATION
DOI 10.1016/j.bspc.2021.103397
DatabaseName CrossRef
DatabaseTitle CrossRef
DatabaseTitleList
DeliveryMethod fulltext_linktorsrc
Discipline Engineering
EISSN 1746-8108
ExternalDocumentID 10_1016_j_bspc_2021_103397
S1746809421009940
GroupedDBID ---
--K
--M
.~1
0R~
1B1
1~.
1~5
23N
4.4
457
4G.
5GY
5VS
6J9
7-5
71M
8P~
AACTN
AAEDT
AAEDW
AAIAV
AAIKJ
AAKOC
AALRI
AAOAW
AAQFI
AAXUO
AAYFN
ABBOA
ABFNM
ABFRF
ABJNI
ABMAC
ABXDB
ABYKQ
ACDAQ
ACGFO
ACGFS
ACNNM
ACRLP
ACZNC
ADBBV
ADEZE
ADMUD
ADTZH
AEBSH
AECPX
AEFWE
AEKER
AENEX
AFKWA
AFTJW
AGHFR
AGUBO
AGYEJ
AHJVU
AHZHX
AIALX
AIEXJ
AIKHN
AITUG
AJBFU
AJOXV
ALMA_UNASSIGNED_HOLDINGS
AMFUW
AMRAJ
AOUOD
AXJTR
BJAXD
BKOJK
BLXMC
CS3
DU5
EBS
EFJIC
EFLBG
EJD
EO8
EO9
EP2
EP3
F5P
FDB
FIRID
FNPLU
FYGXN
G-Q
GBLVA
GBOLZ
HZ~
IHE
J1W
JJJVA
KOM
M41
MO0
N9A
O-L
O9-
OAUVE
OZT
P-8
P-9
P2P
PC.
Q38
RIG
ROL
RPZ
SDF
SDG
SES
SPC
SPCBC
SST
SSV
SSZ
T5K
UNMZH
~G-
9DU
AATTM
AAXKI
AAYWO
AAYXX
ABWVN
ACLOT
ACRPL
ACVFH
ADCNI
ADNMO
AEIPS
AEUPX
AFJKZ
AFPUW
AIGII
AIIUN
AKBMS
AKRWK
AKYEP
ANKPU
APXCP
CITATION
EFKBS
~HD
ID FETCH-LOGICAL-c300t-3f7ea49b6f45c0d47e753d94bb04f5dcc5f13021d42d25d39d51fdf6d29acd473
ISICitedReferencesCount 4
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000782654300004&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 1746-8094
IngestDate Sat Nov 29 07:04:14 EST 2025
Tue Nov 18 22:43:06 EST 2025
Fri Feb 23 02:40:56 EST 2024
IsPeerReviewed true
IsScholarly true
Keywords fMRI
Visual encoding
Shared autoencoder
Alternating optimization
Visual reconstruction
Language English
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-c300t-3f7ea49b6f45c0d47e753d94bb04f5dcc5f13021d42d25d39d51fdf6d29acd473
ParticipantIDs crossref_primary_10_1016_j_bspc_2021_103397
crossref_citationtrail_10_1016_j_bspc_2021_103397
elsevier_sciencedirect_doi_10_1016_j_bspc_2021_103397
PublicationCentury 2000
PublicationDate March 2022
2022-03-00
PublicationDateYYYYMMDD 2022-03-01
PublicationDate_xml – month: 03
  year: 2022
  text: March 2022
PublicationDecade 2020
PublicationTitle Biomedical signal processing and control
PublicationYear 2022
Publisher Elsevier Ltd
Publisher_xml – name: Elsevier Ltd
References VanRullen, R., and Reddy, L. (2018). Reconstructing Faces from fMRI Patterns using Deep Generative Neural Networks.
LeCun, Bengio, Hinton (b0075) 2015; 521
Eickenberg, Gramfort, Varoquaux, Thirion (b0025) 2016; 152
Russakovsky, Deng, Su, Krause, Satheesh, Ma, Huang, Karpathy, Khosla, Bernstein, Berg, Fei-Fei (b0140) 2015; 115
Wen, Shi, Zhang, Lu, Cao, Liu (b0200) 2017
Shen, Dwivedi, Majima, Horikawa, Kamitani (b0150) 2018
Zhang, Y., Wu, Z., Peng, H., and Lin, S. (2020). “A transductive approach for video object segmentation”, in
Horikawa, Kamitani (b0050) 2017; 8
Han, Wen, Shi, Lu, Zhang, Fu, Liu (b0045) 2019; 198
Koch, Zemel, Salakhutdinov (b0065) 2015
Nishimoto, Vu, Naselaris, Benjamini, Yu, Gallant (b0120) 2011; 21
Seeliger, Güçlü, Ambrogioni, Güçlütürk, van Gerven (b0145) 2018; 181
Miyawaki, Uchida, Yamashita, Sato, Morito, Tanabe, Sadato, Kamitani (b0100) 2008; 60
Jiang, Qiao, Wang, Zhang, Chen, Zeng, Bu, Yan (b0060) 2019; 9
Shen, Horikawa, Majima, Kamitani, O’Reilly (b0155) 2019; 15
van Gerven, de Lange, Heskes (b0185) 2010; 22
Guclu, van Gerven (b0035) 2015; 35
Shi, Gong, Ding, Tao, Zheng (b0160) 2018
Chao, Changpinyo, Gong, Sha (b0015) 2016
6949-6958.
Ren, Z., Li, J., Xue, X., Li, X., Yang, F., Jiao, Z., et al. (2019). Reconstructing Perceived Images from Brain Activity by Visually-guided Cognitive Representation and Adversarial Learning.
6707-6717.
Naselaris, Prenger, Kay, Oliver, Gallant (b0115) 2009; 63
Song, Shen, Yang, Liu, Song (b0170) 2018
Norman, Polyn, Detre, Haxby (b0125) 2006; 10
Arjovsky, Chintala, Bottou (b0005) 2017
Chong, Ding, Yan, Pan (b0020) 2020; 408
Zhang, C., Qiao, K., Wang, L., Tong, L., Zeng, Y., and Yan, B. (2018). Constraint-free natural image reconstruction from fMRI signals based on convolutional neural network.
Misra, I., and Maaten, L.v.d. (2020). “Self-supervised learning of pretext-invariant representations”, in
6082-6091.
Styves, Naselaris (b0180) 2017
Mescheder, L., Geiger, A., and Nowozin, S. (2018). Which training methods for GANs do actually converge?
St-Yves, Naselaris (b0175) 2018
Güçlütürk, Güçlü, Seeliger, Bosch, van Lier, van Gerven (b0040) 2017
12.
.
Beliy, Gaziv, Hoogi, Strappini, Golan, Irani (b0010) 2019
Krizhevsky, Sutskever, Hinton (b0070) 2012
Wang, Yao, Kwok, Ni (b0195) 2020; 53
Simonyan, K., and Zisserman, A. (2014). Very deep convolutional networks for large-scale image recognition.
Goodfellow, Pouget-Abadie, Mirza, Xu, Warde-Farley, Ozair (b0030) 2014
Naselaris, Kay, Nishimoto, Gallant (b0110) 2011; 56
Rahman, S., Khan, S., and Barnes, N. (Year). “Transductive learning for zero-shot object detection”, in
Liu, Y., Lee, J., Park, M., Kim, S., Yang, E., Hwang, S.J., et al. (2018). Learning to propagate labels: Transductive propagation network for few-shot learning.
Mirza, M., and Osindero, S. (2014). Conditional generative adversarial nets.
Hughes, Schmitt, Mou, Wang, Zhu (b0055) 2018; 15
Mozafari, M., Reddy, L., and VanRullen, R. (2020). Reconstructing Natural Scenes from fMRI Patterns using BigBiGAN.
Yang, Zhang, Xiang, Torr, Hospedales (b0205) 2018
Wang (10.1016/j.bspc.2021.103397_b0195) 2020; 53
Koch (10.1016/j.bspc.2021.103397_b0065) 2015
Beliy (10.1016/j.bspc.2021.103397_b0010) 2019
Norman (10.1016/j.bspc.2021.103397_b0125) 2006; 10
Russakovsky (10.1016/j.bspc.2021.103397_b0140) 2015; 115
Styves (10.1016/j.bspc.2021.103397_b0180) 2017
Naselaris (10.1016/j.bspc.2021.103397_b0110) 2011; 56
Chong (10.1016/j.bspc.2021.103397_b0020) 2020; 408
Horikawa (10.1016/j.bspc.2021.103397_b0050) 2017; 8
10.1016/j.bspc.2021.103397_b0080
Guclu (10.1016/j.bspc.2021.103397_b0035) 2015; 35
LeCun (10.1016/j.bspc.2021.103397_b0075) 2015; 521
10.1016/j.bspc.2021.103397_b0130
Nishimoto (10.1016/j.bspc.2021.103397_b0120) 2011; 21
Wen (10.1016/j.bspc.2021.103397_b0200) 2017
Arjovsky (10.1016/j.bspc.2021.103397_b0005) 2017
Krizhevsky (10.1016/j.bspc.2021.103397_b0070) 2012
10.1016/j.bspc.2021.103397_b0095
Miyawaki (10.1016/j.bspc.2021.103397_b0100) 2008; 60
Chao (10.1016/j.bspc.2021.103397_b0015) 2016
10.1016/j.bspc.2021.103397_b0215
Shen (10.1016/j.bspc.2021.103397_b0155) 2019; 15
10.1016/j.bspc.2021.103397_b0210
Han (10.1016/j.bspc.2021.103397_b0045) 2019; 198
10.1016/j.bspc.2021.103397_b0135
Eickenberg (10.1016/j.bspc.2021.103397_b0025) 2016; 152
van Gerven (10.1016/j.bspc.2021.103397_b0185) 2010; 22
Seeliger (10.1016/j.bspc.2021.103397_b0145) 2018; 181
St-Yves (10.1016/j.bspc.2021.103397_b0175) 2018
Goodfellow (10.1016/j.bspc.2021.103397_b0030) 2014
10.1016/j.bspc.2021.103397_b0190
10.1016/j.bspc.2021.103397_b0090
Song (10.1016/j.bspc.2021.103397_b0170) 2018
Naselaris (10.1016/j.bspc.2021.103397_b0115) 2009; 63
10.1016/j.bspc.2021.103397_b0085
10.1016/j.bspc.2021.103397_b0165
Shi (10.1016/j.bspc.2021.103397_b0160) 2018
Güçlütürk (10.1016/j.bspc.2021.103397_b0040) 2017
10.1016/j.bspc.2021.103397_b0105
Shen (10.1016/j.bspc.2021.103397_b0150) 2018
Hughes (10.1016/j.bspc.2021.103397_b0055) 2018; 15
Yang (10.1016/j.bspc.2021.103397_b0205) 2018
Jiang (10.1016/j.bspc.2021.103397_b0060) 2019; 9
References_xml – volume: 181
  start-page: 775
  year: 2018
  end-page: 785
  ident: b0145
  article-title: Generative adversarial networks for reconstructing natural images from brain activity
  publication-title: NeuroImage
– volume: 408
  start-page: 216
  year: 2020
  end-page: 230
  ident: b0020
  article-title: Graph-based semi-supervised learning: A review
  publication-title: Neurocomputing
– year: 2015
  ident: b0065
  article-title: Siamese neural networks for one-shot image recognition
  publication-title: in:
– year: 2018
  ident: b0175
  article-title: Generative adversarial networks conditioned on brain activity reconstruct seen images.
  publication-title: bioRxiv
– volume: 63
  start-page: 902
  year: 2009
  end-page: 915
  ident: b0115
  article-title: Bayesian reconstruction of natural images from human brain activity: neuron
  publication-title: Neuron
– start-page: 1
  year: 2017
  end-page: 25
  ident: b0200
  article-title: Neural encoding and decoding with deep learning for dynamic natural vision
  publication-title: Cereb. Cortex
– start-page: 1097
  year: 2012
  end-page: 1105
  ident: b0070
  article-title: ImageNet classification with deep convolutional neural networks
  publication-title: in:
– volume: 115
  start-page: 211
  year: 2015
  end-page: 252
  ident: b0140
  article-title: Imagenet large scale visual recognition challenge
  publication-title: Int. J. Comput. Vision
– reference: Ren, Z., Li, J., Xue, X., Li, X., Yang, F., Jiao, Z., et al. (2019). Reconstructing Perceived Images from Brain Activity by Visually-guided Cognitive Representation and Adversarial Learning.
– year: 2018
  ident: b0150
  article-title: End-to-end deep image reconstruction from human brain activity.
  publication-title: bioRxiv
– reference: Simonyan, K., and Zisserman, A. (2014). Very deep convolutional networks for large-scale image recognition.
– volume: 521
  start-page: 436
  year: 2015
  end-page: 444
  ident: b0075
  article-title: Deep learning
  publication-title: Deep learning.
– reference: 12.
– reference: Zhang, C., Qiao, K., Wang, L., Tong, L., Zeng, Y., and Yan, B. (2018). Constraint-free natural image reconstruction from fMRI signals based on convolutional neural network.
– volume: 53
  start-page: 1
  year: 2020
  end-page: 34
  ident: b0195
  article-title: Generalizing from a few examples: A survey on few-shot learning
  publication-title: ACM Computing Surveys (CSUR)
– reference: Mozafari, M., Reddy, L., and VanRullen, R. (2020). Reconstructing Natural Scenes from fMRI Patterns using BigBiGAN.
– volume: 8
  start-page: 15037
  year: 2017
  ident: b0050
  article-title: Generic decoding of seen and imagined objects using hierarchical visual features
  publication-title: Nat. Commun.
– start-page: 2672
  year: 2014
  end-page: 2680
  ident: b0030
  article-title: Generative adversarial nets
  publication-title: Advances in neural information processing systems
– volume: 56
  start-page: 400
  year: 2011
  end-page: 410
  ident: b0110
  article-title: Encoding and decoding in fMRI
  publication-title: Neuroimage
– reference: Liu, Y., Lee, J., Park, M., Kim, S., Yang, E., Hwang, S.J., et al. (2018). Learning to propagate labels: Transductive propagation network for few-shot learning.
– start-page: 6514
  year: 2019
  end-page: 6524
  ident: b0010
  article-title: From voxels to pixels and back: Self-supervision in natural-image reconstruction from fMRI
  publication-title: Advances in Neural Information Processing Systems
– volume: 60
  start-page: 915
  year: 2008
  end-page: 929
  ident: b0100
  article-title: Visual image reconstruction from human brain activity using a combination of multiscale local image decoders
  publication-title: Neuron
– reference: ), 6082-6091.
– reference: , 6949-6958.
– start-page: 4246
  year: 2017
  end-page: 4257
  ident: b0040
  article-title: Reconstructing perceived faces from brain activations with deep adversarial neural decoding
  publication-title: Adv. Neural Information Processing Systems
– start-page: 52
  year: 2016
  end-page: 68
  ident: b0015
  article-title: An empirical study and analysis of generalized zero-shot learning for object recognition in the wild
  publication-title: : Springer
– reference: , 6707-6717.
– year: 2017
  ident: b0180
  article-title: The feature-weighted receptive field: an interpretable encoding model for complex feature spaces
  publication-title: Neuroimage
– volume: 15
  start-page: e1006633
  year: 2019
  ident: b0155
  article-title: Deep image reconstruction from human brain activity
  publication-title: PLoS Comput. Biol.
– start-page: 299
  year: 2018
  end-page: 315
  ident: b0160
  article-title: Transductive semi-supervised deep learning using min-max features
  publication-title: in:
– volume: 15
  start-page: 784
  year: 2018
  end-page: 788
  ident: b0055
  article-title: Identifying corresponding patches in SAR and optical images with a pseudo-siamese CNN
  publication-title: IEEE Geosci. Remote Sens. Lett.
– reference: Misra, I., and Maaten, L.v.d. (2020). “Self-supervised learning of pretext-invariant representations”, in:
– reference: Mescheder, L., Geiger, A., and Nowozin, S. (2018). Which training methods for GANs do actually converge?
– reference: .
– volume: 10
  start-page: 424
  year: 2006
  end-page: 430
  ident: b0125
  article-title: Beyond mind-reading: multi-voxel pattern analysis of fMRI data
  publication-title: Trends in cognitive sciences
– reference: Zhang, Y., Wu, Z., Peng, H., and Lin, S. (2020). “A transductive approach for video object segmentation”, in:
– volume: 21
  start-page: 1641
  year: 2011
  end-page: 1646
  ident: b0120
  article-title: Reconstructing visual experiences from brain activity evoked by natural movies
  publication-title: Curr. Biol.
– reference: Rahman, S., Khan, S., and Barnes, N. (Year). “Transductive learning for zero-shot object detection”, in:
– volume: 9
  start-page: 4749
  year: 2019
  ident: b0060
  article-title: Siamese reconstruction network: accurate image reconstruction from human brain activity by learning to compare
  publication-title: Appl. Sci.
– volume: 198
  start-page: 125
  year: 2019
  end-page: 136
  ident: b0045
  article-title: Variational autoencoder: An unsupervised model for encoding and decoding fMRI activity in visual cortex
  publication-title: NeuroImage
– reference: Mirza, M., and Osindero, S. (2014). Conditional generative adversarial nets.
– start-page: 1024
  year: 2018
  end-page: 1033
  ident: b0170
  article-title: Transductive unbiased embedding for zero-shot learning
  publication-title: in:
– reference: VanRullen, R., and Reddy, L. (2018). Reconstructing Faces from fMRI Patterns using Deep Generative Neural Networks.
– start-page: 214
  year: 2017
  end-page: 223
  ident: b0005
  article-title: Wasserstein generative adversarial networks
  publication-title: International Conference on Machine Learning
– volume: 152
  year: 2016
  ident: b0025
  article-title: Seeing it all: Convolutional network layers map the function of the human visual system
  publication-title: Neuroimage
– volume: 22
  start-page: 3127
  year: 2010
  end-page: 3142
  ident: b0185
  article-title: Neural decoding with hierarchical generative models
  publication-title: Neural Comput.
– year: 2018
  ident: b0205
  article-title: Learning to compare
– volume: 35
  start-page: 10005
  year: 2015
  end-page: 10014
  ident: b0035
  article-title: Deep neural networks reveal a gradient in the complexity of neural representations across the ventral stream
  publication-title: J. Neurosci.
– ident: 10.1016/j.bspc.2021.103397_b0130
– volume: 15
  start-page: e1006633
  issue: 1
  year: 2019
  ident: 10.1016/j.bspc.2021.103397_b0155
  article-title: Deep image reconstruction from human brain activity
  publication-title: PLoS Comput. Biol.
  doi: 10.1371/journal.pcbi.1006633
– ident: 10.1016/j.bspc.2021.103397_b0215
  doi: 10.1109/CVPR42600.2020.00698
– ident: 10.1016/j.bspc.2021.103397_b0090
– volume: 115
  start-page: 211
  issue: 3
  year: 2015
  ident: 10.1016/j.bspc.2021.103397_b0140
  article-title: Imagenet large scale visual recognition challenge
  publication-title: Int. J. Comput. Vision
  doi: 10.1007/s11263-015-0816-y
– volume: 22
  start-page: 3127
  issue: 12
  year: 2010
  ident: 10.1016/j.bspc.2021.103397_b0185
  article-title: Neural decoding with hierarchical generative models
  publication-title: Neural Comput.
  doi: 10.1162/NECO_a_00047
– volume: 9
  start-page: 4749
  issue: 22
  year: 2019
  ident: 10.1016/j.bspc.2021.103397_b0060
  article-title: Siamese reconstruction network: accurate image reconstruction from human brain activity by learning to compare
  publication-title: Appl. Sci.
  doi: 10.3390/app9224749
– volume: 408
  start-page: 216
  year: 2020
  ident: 10.1016/j.bspc.2021.103397_b0020
  article-title: Graph-based semi-supervised learning: A review
  publication-title: Neurocomputing
  doi: 10.1016/j.neucom.2019.12.130
– volume: 181
  start-page: 775
  year: 2018
  ident: 10.1016/j.bspc.2021.103397_b0145
  article-title: Generative adversarial networks for reconstructing natural images from brain activity
  publication-title: NeuroImage
  doi: 10.1016/j.neuroimage.2018.07.043
– volume: 60
  start-page: 915
  issue: 5
  year: 2008
  ident: 10.1016/j.bspc.2021.103397_b0100
  article-title: Visual image reconstruction from human brain activity using a combination of multiscale local image decoders
  publication-title: Neuron
  doi: 10.1016/j.neuron.2008.11.004
– ident: 10.1016/j.bspc.2021.103397_b0165
– year: 2017
  ident: 10.1016/j.bspc.2021.103397_b0180
  article-title: The feature-weighted receptive field: an interpretable encoding model for complex feature spaces
  publication-title: Neuroimage
– start-page: 6514
  year: 2019
  ident: 10.1016/j.bspc.2021.103397_b0010
  article-title: From voxels to pixels and back: Self-supervision in natural-image reconstruction from fMRI
  publication-title: Advances in Neural Information Processing Systems
– ident: 10.1016/j.bspc.2021.103397_b0190
  doi: 10.32470/CCN.2018.1031-0
– start-page: 214
  year: 2017
  ident: 10.1016/j.bspc.2021.103397_b0005
  article-title: Wasserstein generative adversarial networks
  publication-title: International Conference on Machine Learning
– ident: 10.1016/j.bspc.2021.103397_b0085
– volume: 521
  start-page: 436
  issue: 7553
  year: 2015
  ident: 10.1016/j.bspc.2021.103397_b0075
  article-title: Deep learning
  publication-title: Deep learning. nature
– year: 2018
  ident: 10.1016/j.bspc.2021.103397_b0205
– start-page: 1024
  year: 2018
  ident: 10.1016/j.bspc.2021.103397_b0170
  article-title: Transductive unbiased embedding for zero-shot learning
– volume: 21
  start-page: 1641
  issue: 19
  year: 2011
  ident: 10.1016/j.bspc.2021.103397_b0120
  article-title: Reconstructing visual experiences from brain activity evoked by natural movies
  publication-title: Curr. Biol.
  doi: 10.1016/j.cub.2011.08.031
– volume: 35
  start-page: 10005
  issue: 27
  year: 2015
  ident: 10.1016/j.bspc.2021.103397_b0035
  article-title: Deep neural networks reveal a gradient in the complexity of neural representations across the ventral stream
  publication-title: J. Neurosci.
  doi: 10.1523/JNEUROSCI.5023-14.2015
– volume: 15
  start-page: 784
  issue: 5
  year: 2018
  ident: 10.1016/j.bspc.2021.103397_b0055
  article-title: Identifying corresponding patches in SAR and optical images with a pseudo-siamese CNN
  publication-title: IEEE Geosci. Remote Sens. Lett.
  doi: 10.1109/LGRS.2018.2799232
– start-page: 2672
  year: 2014
  ident: 10.1016/j.bspc.2021.103397_b0030
  article-title: Generative adversarial nets
  publication-title: Advances in neural information processing systems
– year: 2018
  ident: 10.1016/j.bspc.2021.103397_b0175
  article-title: Generative adversarial networks conditioned on brain activity reconstruct seen images.
  publication-title: bioRxiv
– volume: 56
  start-page: 400
  issue: 2
  year: 2011
  ident: 10.1016/j.bspc.2021.103397_b0110
  article-title: Encoding and decoding in fMRI
  publication-title: Neuroimage
  doi: 10.1016/j.neuroimage.2010.07.073
– start-page: 4246
  year: 2017
  ident: 10.1016/j.bspc.2021.103397_b0040
  article-title: Reconstructing perceived faces from brain activations with deep adversarial neural decoding
  publication-title: Adv. Neural Information Processing Systems
– start-page: 1
  year: 2017
  ident: 10.1016/j.bspc.2021.103397_b0200
  article-title: Neural encoding and decoding with deep learning for dynamic natural vision
  publication-title: Cereb. Cortex
– year: 2015
  ident: 10.1016/j.bspc.2021.103397_b0065
  article-title: Siamese neural networks for one-shot image recognition
– ident: 10.1016/j.bspc.2021.103397_b0105
  doi: 10.1109/IJCNN48605.2020.9206960
– volume: 63
  start-page: 902
  issue: 6
  year: 2009
  ident: 10.1016/j.bspc.2021.103397_b0115
  article-title: Bayesian reconstruction of natural images from human brain activity: neuron
  publication-title: Neuron
  doi: 10.1016/j.neuron.2009.09.006
– start-page: 299
  year: 2018
  ident: 10.1016/j.bspc.2021.103397_b0160
  article-title: Transductive semi-supervised deep learning using min-max features
– ident: 10.1016/j.bspc.2021.103397_b0135
– volume: 8
  start-page: 15037
  year: 2017
  ident: 10.1016/j.bspc.2021.103397_b0050
  article-title: Generic decoding of seen and imagined objects using hierarchical visual features
  publication-title: Nat. Commun.
  doi: 10.1038/ncomms15037
– volume: 198
  start-page: 125
  year: 2019
  ident: 10.1016/j.bspc.2021.103397_b0045
  article-title: Variational autoencoder: An unsupervised model for encoding and decoding fMRI activity in visual cortex
  publication-title: NeuroImage
  doi: 10.1016/j.neuroimage.2019.05.039
– start-page: 1097
  year: 2012
  ident: 10.1016/j.bspc.2021.103397_b0070
  article-title: ImageNet classification with deep convolutional neural networks
– ident: 10.1016/j.bspc.2021.103397_b0080
– volume: 152
  year: 2016
  ident: 10.1016/j.bspc.2021.103397_b0025
  article-title: Seeing it all: Convolutional network layers map the function of the human visual system
  publication-title: Neuroimage
– start-page: 52
  year: 2016
  ident: 10.1016/j.bspc.2021.103397_b0015
  article-title: An empirical study and analysis of generalized zero-shot learning for object recognition in the wild
  publication-title: European conference on computer vision: Springer
– volume: 53
  start-page: 1
  issue: 3
  year: 2020
  ident: 10.1016/j.bspc.2021.103397_b0195
  article-title: Generalizing from a few examples: A survey on few-shot learning
  publication-title: ACM Computing Surveys (CSUR)
  doi: 10.1145/3386252
– year: 2018
  ident: 10.1016/j.bspc.2021.103397_b0150
  article-title: End-to-end deep image reconstruction from human brain activity.
  publication-title: bioRxiv
– ident: 10.1016/j.bspc.2021.103397_b0210
  doi: 10.3389/fnhum.2018.00242
– ident: 10.1016/j.bspc.2021.103397_b0095
  doi: 10.1109/CVPR42600.2020.00674
– volume: 10
  start-page: 424
  issue: 9
  year: 2006
  ident: 10.1016/j.bspc.2021.103397_b0125
  article-title: Beyond mind-reading: multi-voxel pattern analysis of fMRI data
  publication-title: Trends in cognitive sciences
  doi: 10.1016/j.tics.2006.07.005
SSID ssj0048714
Score 2.2853737
Snippet •We proposed the mutual promotion of visual encoding and reconstruction models.•We designed alternating optimization bassed on seim-supervised learning.•We...
SourceID crossref
elsevier
SourceType Enrichment Source
Index Database
Publisher
StartPage 103397
SubjectTerms Alternating optimization
fMRI
Shared autoencoder
Visual encoding
Visual reconstruction
Title Reconstructing natural images from human fMRI by alternating encoding and decoding with shared autoencoder regularization
URI https://dx.doi.org/10.1016/j.bspc.2021.103397
Volume 73
WOSCitedRecordID wos000782654300004&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVESC
  databaseName: Elsevier SD Freedom Collection Journals 2021
  customDbUrl:
  eissn: 1746-8108
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0048714
  issn: 1746-8094
  databaseCode: AIEXJ
  dateStart: 20060101
  isFulltext: true
  titleUrlDefault: https://www.sciencedirect.com
  providerName: Elsevier
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV1LT9tAEF6loYdyqKAPQVuqPfQWGdne9euIEFWpCkIVVXOz7H2AEThRCIj8lf7azuzDcWiL4NCLlax211bmy3hmduYbQj6lAowKkfKApVwEXGodFDqvg0wBWmpkZ8lNofC37Pg4H4-Lk8Hgl6-Fub3M2ja_uyum_1XUMAbCxtLZJ4i72xQG4DMIHa4gdrg-SvDoUHpa2PZsZJg7kVnjqkI6B1NOYhvz6aPvh2h9mgNzDArCbKS1lL5uUSr3xeaxn5tc9epmPjGz1Gw0M43sZ66Uc-V82FT125LL5gwt3qmtSPB7uxT5Lu7aVBOb3NEsEw5c3UgPwD9ddBs86MXNcriLeu-fN_0wBnjAXR6X07wZR2Zk2_HYq-aM9XRrFDJmc3n_UPs2AnGxW19PkZYyjnaXk1c5tu-9-7qMRJ_sdlHiHiXuUdo9npG1OEuKfEjW9g4Pxl_9ex48PcMc3z24K8my2YP3n-TvZk_PlDndIC-dD0L3LHY2yUC1r8h6j5nyNVmsoog6FFGLIoooogZFFFFE6wXtoYh6FFGQNPUooogialFEeyiiqyh6Q358Pjjd_xK4Lh2BYGE4D5jOVMWLOtU8EaHkmQIPWBa8rkOuEylEovFwPJI8lnEiWSGTSEudyrioBExnb8mwnbRqi9BcREzFNQP9kcA-VR4xDQsltggArRFuk8j_iKVwFPbYSeWy_Lf4tsmoWzO1BC4Pzk68bEpnglrTsgSoPbDu3ZPu8p68WP4FPpAhCFPtkOfidt5czz46nP0GEgetFg
linkProvider Elsevier
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Reconstructing+natural+images+from+human+fMRI+by+alternating+encoding+and+decoding+with+shared+autoencoder+regularization&rft.jtitle=Biomedical+signal+processing+and+control&rft.au=Qiao%2C+Kai&rft.au=Chen%2C+Jian&rft.au=Wang%2C+Linyuan&rft.au=Zhang%2C+Chi&rft.date=2022-03-01&rft.issn=1746-8094&rft.volume=73&rft.spage=103397&rft_id=info:doi/10.1016%2Fj.bspc.2021.103397&rft.externalDBID=n%2Fa&rft.externalDocID=10_1016_j_bspc_2021_103397
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1746-8094&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1746-8094&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1746-8094&client=summon