DMD training framework with Imagenet pretrained networks
The Driver Monitoring Dataset is the largest visual dataset for real driving actions, with footage from synchronized multiple cameras (body, face, hands) and multiple streams (RGB, Depth, IR) recorded in two scenarios (real car, driving simulator). Different annotated labels related to distraction, fatigue and gaze-head pose can be used to train Deep Learning models for Driver Monitor Systems.
We used dataset with 11 kinds of labels
- Change Gear
- Drinking
- Hair and Makeup
- Phone Call
- Operating Radio
- Reach Backseat
- Reach Side
- Stand Still or Waiting
- Talking to Passenger
- Texting
- Safe Driving