Ubiquitous Traffic Eyes: trajectory dataset focus on multiple traffic states and state transition on urban expressways

Vehicle trajectory data provides critical information for traffic flow modeling and analysis during different traffic states and state transitions. This paper introduces a trajectory dataset based on aerial videos called Ubiquitous Traffic Eyes (UTE). It contains high-resolution vehicle trajectories automatically extracted from UAV videos, addressing issues such as pixel shake, vehicle detection, and trajectory fragment reconstruction. To bridge the gap in available data for traffic transition conditions, UTE captures data from typical state transition periods and fixed bottleneck locations, such as interweave areas. The current dataset includes data from seven distinct locations, encompassing freeway basic segments, merge/diverge segments, and weaving segments. The dataset presents trajectory data under diverse traffic conditions, including free flow, congestion, and transitions. This facilitates a comprehensive analysis of how traffic evolves and adapts, particularly in regions characterized by frequent maneuvers. The dataset is available at http://seutraffic.com/.