Urban Place Recognition Dataset

This dataset provides visual and inertial information as well as manually-​annotated ground-​truth for sequences specifically captured for place recognition tasks, outdoors in urban scenes. The recordings were made with a side-​looking camera using hand-​held setups at different heights, as well as with the help of a drone, visiting the same scene at different times of the day and the year. As a result, this dataset poses significant challenges in viewpoint, illumination and situational changes.

All the sequences were recorded using a high quality visual-​inertial sensor providing monocular, grayscale, global-​shutter images at 20 Hz and time-​synchronized inertial measurements. This dataset is freeley available and can be downloaded from here.

Users of this dataset are asked to cite the following paper, where it was originally presented:

Fabiola Maffra, Zetao Chen and Margarita Chli, “Viewpoint-​tolerant Place Recognition combining 2D and 3D information for UAV navigation”, in Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), 2018. DOI E-​citations