logo
CanyamAI summaries for academic research
language
MobileMobile
  • Phone FeatureWeChat Mini Program
EnglishEnglish
noticeNotice
BlogBlog
delete
arrow
Return

MobileFold: Linear-Complexity Global Modeling via Spatial Fold Convolution

delete2026-02-01
delete0
PRE
AI
A
Ang Li *
W
Wei Lu
DOI：10.1016/j.patrec.2026.01.029delete
deleteOriginal
deleteOriginal request for help
deleteShare
deleteSave
Abstract

Abstract

En 中文
• Spatial Fold Convolution transforms global modeling into channel-wise operations with linear complexity • Global-Local Selective Fusion reduces redundancy between local and global features • MobileFold-Tiny outperforms MobileViTV3 with 30% fewer computations • MobileFold demonstrates strong generalization across both classification and detection tasks

Journal

Pattern Recognition Letters cover
Pattern Recognition Letters
IF:
3.3
Papers:
7.9K
Citations:
1.6W

Organization

No organization information available