A one-stage model with channel subgroup patch embedding and subgroup-masked self-supervised learning improves RF-based multi-person pose estimation, beating prior two-stage methods with 98% fewer parameters.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Self-supervised One-Stage Learning for RF-based Multi-Person Pose Estimation
A one-stage model with channel subgroup patch embedding and subgroup-masked self-supervised learning improves RF-based multi-person pose estimation, beating prior two-stage methods with 98% fewer parameters.