A Survey on Path Prediction Techniques for Vulnerable Road Users: From Traditional to Deep-Learning Approaches

Behavior analysis of Vulnerable Road Users (VRU)s has become a crucial topic in the computer vision research area. In recent decades, numerous papers have extensively addressed the problem of VRU path prediction, which has a wide range of applications such as video surveillance and autonomous driving. The behavioral complexity of VRUs has forced researchers to employ various techniques in order to develop more comprehensive models that potentially respond better to VRUs movement patterns. This indeed has led to development of a large variety of models and approaches in the literature. The aim of this paper is to a) provide a comprehensive review of developed path prediction methods, b) individuate and classify the proposed methods from multiple viewpoints, and c) present a framework for better understanding of various aspects in VRUs path prediction problems.