.png){kind=icon}
.png){kind=icon}
Return
Towards generalizable driver drowsiness detection: A unified framework with geometric transformations and self-correcting adversarial learning
.png){kind=icon}
PRE
AI
DOI：10.1016/j.eswa.2026.132423.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• The first video-based cross-domain driver drowsiness detection. • A novel multi-view facial landmark augmentation strategy is proposed. • An adversarial domain module to learn domain-invariant features. • A generalizable loss function for label noise and inconsistencies.
Keywords:
driver drowsiness detection
cross-domain learning
facial landmark augmentation
adversarial domain module
generalizable loss function
Journal
IF:
7.5
Papers:
2.9W
Citations:
10.2W