Argoverse is a large-scale autonomous driving dataset that includes 3D tracking and motion forecasting data with rich maps. The dataset was collected using a fleet of autonomous vehicles in Pittsburgh and Miami. It includes 360° images from 7 cameras, 3D point clouds from long-range LiDAR, 6-DOF pose, and 3D track annotations. The Argoverse 3D Tracking dataset includes over 300,000 5-second tracked scenarios with a particular vehicle identified for trajectory forecasting. The dataset also includes HD maps with 290 km of mapped lanes, providing geometric and semantic metadata. All data is released under a Creative Commons license. The dataset includes detailed map information such as lane direction, driveable area, and ground height, which improves the accuracy of 3D object tracking and motion forecasting. The dataset is the first to include HD maps with semantic vector maps of road infrastructure and traffic rules. The dataset also includes a large-scale forecasting dataset consisting of trajectory data for interesting scenarios such as turns at intersections, high traffic clutter, and lane changes. The dataset includes a map API that can be used to develop map-based perception and forecasting algorithms. The dataset is the first self-driving vehicle dataset with a semantic vector map of road infrastructure and traffic rules. The inclusion of HD map information also means our dataset is the first large-scale benchmark for automatic map creation. The dataset includes a large-scale forecasting dataset consisting of trajectory data for interesting scenarios such as turns at intersections, high traffic clutter, and lane changes. The dataset includes a map API that can be used to develop map-based perception and forecasting algorithms. The dataset is the first self-driving vehicle dataset with a semantic vector map of road infrastructure and traffic rules. The inclusion of HD map information also means our dataset is the first large-scale benchmark for automatic map creation. The dataset includes a large-scale forecasting dataset consisting of trajectory data for interesting scenarios such as turns at intersections, high traffic clutter, and lane changes. The dataset includes a map API that can be used to develop map-based perception and forecasting algorithms. The dataset is the first self-driving vehicle dataset with a semantic vector map of road infrastructure and traffic rules. The inclusion of HD map information also means our dataset is the first large-scale benchmark for automatic map creation.Argoverse is a large-scale autonomous driving dataset that includes 3D tracking and motion forecasting data with rich maps. The dataset was collected using a fleet of autonomous vehicles in Pittsburgh and Miami. It includes 360° images from 7 cameras, 3D point clouds from long-range LiDAR, 6-DOF pose, and 3D track annotations. The Argoverse 3D Tracking dataset includes over 300,000 5-second tracked scenarios with a particular vehicle identified for trajectory forecasting. The dataset also includes HD maps with 290 km of mapped lanes, providing geometric and semantic metadata. All data is released under a Creative Commons license. The dataset includes detailed map information such as lane direction, driveable area, and ground height, which improves the accuracy of 3D object tracking and motion forecasting. The dataset is the first to include HD maps with semantic vector maps of road infrastructure and traffic rules. The dataset also includes a large-scale forecasting dataset consisting of trajectory data for interesting scenarios such as turns at intersections, high traffic clutter, and lane changes. The dataset includes a map API that can be used to develop map-based perception and forecasting algorithms. The dataset is the first self-driving vehicle dataset with a semantic vector map of road infrastructure and traffic rules. The inclusion of HD map information also means our dataset is the first large-scale benchmark for automatic map creation. The dataset includes a large-scale forecasting dataset consisting of trajectory data for interesting scenarios such as turns at intersections, high traffic clutter, and lane changes. The dataset includes a map API that can be used to develop map-based perception and forecasting algorithms. The dataset is the first self-driving vehicle dataset with a semantic vector map of road infrastructure and traffic rules. The inclusion of HD map information also means our dataset is the first large-scale benchmark for automatic map creation. The dataset includes a large-scale forecasting dataset consisting of trajectory data for interesting scenarios such as turns at intersections, high traffic clutter, and lane changes. The dataset includes a map API that can be used to develop map-based perception and forecasting algorithms. The dataset is the first self-driving vehicle dataset with a semantic vector map of road infrastructure and traffic rules. The inclusion of HD map information also means our dataset is the first large-scale benchmark for automatic map creation.