Driver fatigue is one of the main causes of road accidents, especially at long distances and at night. driving. In this proposed research, we introduce a real-time vision-based system for detecting drowsiness based on Mediapipe and OpenCV for Facial Landmark detection. The system calculates three metrics, namely: Eye Aspect Ratio Eye Aspect Ratio (EAR), Mouth Aspect Ratio (MAR) and head tilt--to identify early onset of drowsiness (eye closure for a prolonged time) yawning, and nodding. Intersection over Union In this phase calibration is performed by the individual driver thresholds. (IoU) filtering is implemented to ensure that the driver's face is only monitored in multi person scenarios. Visual fatigue: Upon seeing fatigue and sound alarms are generated to avoid possible accidents. Compared to the existing system, the proposed system is non-intrusive, and has low cost.