Recognition of Facial Expression and Drowsiness Using Landmarks

Emotions are important for extracting facial expressions and they can be calculated by still images, Video frames. While driving the driver should be alert because many accidents occur due to the drowsiness of the driver. To overcome this, we can detect drowsiness and alert the driver. In this paper, we have calculated the drowsiness and facial expression by using facial landmarks with the Euclidean distance algorithm. The landmarks detection is done with the shape-predictor file which is trained with the IBUG 300-W dataset in which about 300 facial expressions are recorded. The shape predictor file is to detect the faces and marks points. Through the shape predictor, we can detect multiple faces that may not possible by neural networks.