MSTFormer: multi-granularity spatial-temporal transformers for 3D human pose estimation

Abstract The 2D-to-3D lifting approach based on multi-granularity methods effectively captures spatial-temporal features at various scales. Existing multi-granularity methods primarily focus on extracting joint features through graph-based approaches, which abstract information at different levels....

Full description

Bibliographic Details

Published in:	Journal of King Saud University: Computer and Information Sciences
Main Authors:	Hao Lin, Sheng Xu, Chengyue Su
Format:	Article
Language:	English
Published:	Springer 2025-04-01
Subjects:	3D human pose estimation Multi-granularity method Skeleton structured information
Online Access:	https://doi.org/10.1007/s44443-025-00023-4