4 datasets found

The dataset is a large-scale ground-to-2.5D map geolocalization dataset, consisting of 113767 panoramic images and geo-tagged multi-modal maps, covering multiple cities.

This study proposes a framework for developing a novel deep learning-based map-matching model in the limited ground-truth data environment.

The dataset used in this paper is a digital road network dataset, consisting of more than 8500 roads, used for map-matching.

The dataset used in this paper is a GPS trajectory dataset collected from a vehicle in Seattle, with 7351 GPS records, including latitude and longitude with a time stamp.