YawDD+ frame-level annotations improve yawn classification to 99.34% accuracy and detection to 95.69% mAP on Jetson hardware compared to video-level labels.
Two-stream spatial–temporal graph convolutional networks for driver drowsiness detection.IEEE Transactions on Cybernetics, 52(12):13821–13833
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
-
YawDD+: Frame-level Annotations for Accurate Yawn Prediction
YawDD+ frame-level annotations improve yawn classification to 99.34% accuracy and detection to 95.69% mAP on Jetson hardware compared to video-level labels.