A modular ISLR training recipe combining speed and quality augmentations with sign boundary regression and IoU-weighted cross-entropy improves recognition accuracy on WLASL, AUTSL, Slovo, and the new SlovoExt corpus.
Randaugment: Practical automated data augmen- tation with a reduced search space
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Training Strategies for Isolated Sign Language Recognition
A modular ISLR training recipe combining speed and quality augmentations with sign boundary regression and IoU-weighted cross-entropy improves recognition accuracy on WLASL, AUTSL, Slovo, and the new SlovoExt corpus.