Research on Assessment of E-learning Efficiency Based on Facial Direction

Xiao-meng FAN, Wei ZHANG, Yi-jun WANG


As the e-learning is becoming more and more popular and widely used all over the world, finding a reliable and effecitve method to evaluate performance of leanrer becomes a problem. However, most of current methods fail to reflect the true study status of the learner. In this paper, an image processing technique based method to obtain index of distraction by estimating head pose of learner is proposed. It uses a single PC camera to capture the facial information of the learner and then gets positions of facial features by analyzing the facial image. Then it uses position information to estimate the roll angle of the head and the horizontal yaw angle of the face in order to calculate the key index to judge the distraction rate of the learner with a designated formula. A series of experiments show that despite the accuracy of head pose estimation is not so well (about 40%), the success rate of judging distraction status can reach nearly 80% if the head pose estimation succeed.


Face recognition, Head pose estimation, Skin detection



  • There are currently no refbacks.