Driver Drowsiness Detection Model Video Dataset Collection

Client's Challenge & Our Solution

A leading automotive company aimed to develop a driver drowsiness detection model and required a diverse and comprehensive video dataset. The client needed videos of drivers engaged in various activities such as enjoying music, interacting with the infotainment system, drinking from bottles and cups, talking with co-passengers, wearing glasses, and feeling sleepy. These videos needed to be captured from both front and side angles and should feature participants from multiple ethnic backgrounds, including East Asian, South Asian, Middle Eastern, Caucasian, and Hispanic.

FutureBeeAI provided a full-service video data collection solution, sourcing videos from 1000 participants from diverse demographics across various angles. We ensured each video met the client’s specifications for activity labeling and video diversity, delivering the dataset within 6 weeks.

Outcome & Features:

Collected videos from 1000 participants featuring diverse activities related to driver behavior, including activities like drinking, exhibiting signs of drowsiness, and more

Captured the videos from both front and side angles, ensuring diverse perspectives to train a robust model.

Delivered a diverse dataset featuring multi-ethnicity participants, with accurate activity classification, completed within 6 weeks.