Lightweight Pixel Difference Networks for Efficient Visual Representation Learning

Approved

Classifications

MinEdu publication type
A1 Journal article (peer-reviewed)
Definition
Article
Target group
Scientific
Peer reviewed
Peer-reviewed
Article type
Journal article
Host publication type
Journal

Authors of the publication

Number of authors
7
Authors
Su, Zhuo; Zhang, Jiehua; Wang, Longguang; Zhang, Hua; Liu, Zhen; Pietikäinen, Matti; Liu, Li

Publication channel information

Title of journal/series
IEEE transactions on pattern analysis and machine intelligence
ISSN (print)
0162-8828
ISSN (electronic)
1939-3539
ISSN (linking)
0162-8828
Publisher
IEEE
Publication forum ID
57568
Publication forum level
3
Country of publication
United States
Internationality
Yes

Detailed publication information

Publication year
2023
Bibliographical publication year
2023
Reporting year
2023
Journal/series volume number
45
Journal/series issue number
12
Page numbers
14956-14974
DOI
10.1109/TPAMI.2023.3300513
Language of publication
English

Co-publication information

International co-publication
Yes
Co-publication with a company
No

Availability

Classification and additional information

MinEdu field of science classification
213 Electronic, automation and communications engineering, electronics
Keywords
Binary neural networks; convolutional neural networks; edge detection; efficient representation learning; facial recognition; image classification

Funding information

Funding information in the publication
This work was supported in part by the National Key Research and Development Program of China under Grant 2021YFB3100800, in part by the Academy of Finland under Grant 331883, in part by the Infotech Project FRAGES, in part by the National Natural Science Foundation of China under Grants 62376283, 61872379, and 62022091, and in part by the CSC IT Center for Science, Finland for computational resources.
Funders
Funder
Research Council of Finland (former Academy of Finland)
Name of funding
-
Funding decision
-

Source database ID

WoS ID
WOS:001104973300056
Scopus ID
2-s2.0-85166767649