Driving Behavior Recognition Method Based on Tutor-Student Network

This paper presents a driving behavior recognition model based on tutor-student network. Considering that driving behavior occurs in a local area, this paper divides the task of driving behavior recognition into two subtasks: action location and action classification. Aiming at the task of action location, a tutor network with shallow network layer receiving high-resolution image input is designed. The tutor network weakens the action area according to the response of feature map. On the basis of action location and action classification task, a student network with deeper network layer is designed to receive the input of low-resolution action area image. High-level semantic features which arc extracted from student network arc used to achieve high accuracy classification. Experimental results show that the tutor-student network model can bring high recognition accuracy and strong robustness.