Local Shuffled Skeleton Position Embedding Vision Transformer for human activity recognition

Vision Transformers (ViTs) in human activity recognition tasks suffer from inadequate spatial modeling through conventional position embeddings, leading to over-reliance on fixed positional information. This paper proposes Shuffled Positional Embedding (SPE), a mechanism that randomly disrupts the o...

Full description

Bibliographic Details

Main Authors:	Zihui Yan, Xiyu Shi, Varuna De-Silva
Format:	Default Conference proceeding
Published:	2025
Subjects:	Shuffled position embedding Vision transformer Human activity recognition
Online Access:	https://hdl.handle.net/2134/30601004.v1
Tags:	Add Tag No Tags, Be the first to tag this record!