Detection of Cyclists' Crossing Intentions for Autonomous Vehicles

Abstract

Improving the safety of bicycle riders is one of the critical issues for Autonomous Driving. The crossing intention of the cyclist is expected to be predicted from the onboard camera of autonomous vehicle. In a real traffic situation, a cyclist usually turns his or her head to check the situation of the back of him or her before he or she crosses the road. Therefore, the action of turning head is an important signal to indicate the intention of crossing a road. This paper proposes to detect the behavior of the turning head based on the body and head orientation using deep neural networks. The proposed system firstly detects the cyclists and extracts the area of the cyclist based on a segmentation neural network. After that, the image of each cyclist is processed by a pose estimation neural network to detect each joint of the cyclist. Finally, the segmented area of the cyclist and the heatmap of each joint of the cyclist are imported into a classification neural network to estimate the body and head orientation, and further predict the crossing intention of the cyclist. A series of experiments have been performed and the experimental results show that the proposed system has a satisfactory performance compared to the conventional method.