Ghost shuffle lightweight pose network with effective feature representation and learning for human pose estimation

Ghost shuffle lightweight pose network with effective feature representation and learning for human pose estimation

Abstract Despite their success, existing human pose estimation approaches mostly have complex architectures, high cost, and lack of lightweight modules. To address this problem, this paper proposes a Ghost Shuffle Lightweight Pose Network (GSLPN) with a more lightweight and efficient network archite...

Full description

Bibliographic Details
Main Authors: Senquan Yang, Jiajun Wen, Junjun Fan
Format: Article
Language:English
Published: Wiley 2022-09-01
Series:IET Computer Vision
Subjects:
ghost shuffle lightweight module
human pose estimation
lightweight pose parallel attention
Online Access:https://doi.org/10.1049/cvi2.12110

Internet

https://doi.org/10.1049/cvi2.12110

Similar Items