EMDUL expands mmWave HPE datasets via pseudo-labeling of unlabeled data and a closed-form LiDAR-to-mmWave converter, reducing pose estimation errors by 15.1% in-domain and 18.9% out-of-domain.
Pseudo-Label : The Simple and Efficient Semi-Supervised Learning Method for Deep Neural Net- works
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Expanding mmWave Datasets for Human Pose Estimation with Unlabeled Data and LiDAR Datasets
EMDUL expands mmWave HPE datasets via pseudo-labeling of unlabeled data and a closed-form LiDAR-to-mmWave converter, reducing pose estimation errors by 15.1% in-domain and 18.9% out-of-domain.