MC-MLP: A Multiple Coordinate Frames MLP-Like Architecture for Vision

Zhimin Zhu, Jianguo Zhao, Tong Mu, Yuliang Yang*, Mengyu Zhu

*此作品的通讯作者

科研成果: 书/报告/会议事项章节会议稿件同行评审

摘要

In deep learning, Multi-Layer Perceptrons (MLPs) have once again garnered attention from researchers. This paper introduces MC-MLP, a general MLP-like backbone for computer vision that is composed of a series of fully-connected (FC) layers. In MC-MLP, we propose that the same semantic information has varying levels of difficulty in learning, depending on the coordinate frame of features. To address this, we perform an orthogonal transform on the feature information, equivalent to changing the coordinate frame of features. Through this design, MC-MLP is equipped with multi-coordinate frame receptive fields and the ability to learn information across different coordinate frames. Experiments demonstrate that MC-MLP outperforms most MLPs in image classification tasks, achieving better performance at the same parameter level. The code will be available at: https://github.com/ZZM11/MC-MLP.

源语言英语
主期刊名Artificial Neural Networks and Machine Learning – ICANN 2023 - 32nd International Conference on Artificial Neural Networks, Proceedings
编辑Lazaros Iliadis, Antonios Papaleonidas, Plamen Angelov, Chrisina Jayne
出版商Springer Science and Business Media Deutschland GmbH
446-456
页数11
ISBN(印刷版)9783031442094
DOI
出版状态已出版 - 2023
活动32nd International Conference on Artificial Neural Networks, ICANN 2023 - Heraklion, 希腊
期限: 26 9月 202329 9月 2023

出版系列

姓名Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
14255 LNCS
ISSN(印刷版)0302-9743
ISSN(电子版)1611-3349

会议

会议32nd International Conference on Artificial Neural Networks, ICANN 2023
国家/地区希腊
Heraklion
时期26/09/2329/09/23

指纹

探究 'MC-MLP: A Multiple Coordinate Frames MLP-Like Architecture for Vision' 的科研主题。它们共同构成独一无二的指纹。

引用此