THÖR: Tracking Human Motion at Örebro University

  • Andrey Rudenko (Creator)
  • Tomasz Kucner (Creator)
  • Chittaranjan Srinivas Swaminathan (Creator)
  • Ravi T. Chadalavada (Creator)
  • Kai O. Arras (Creator)
  • Achim J. Lilienthal (Creator)



THÖR: a dataset of motion trajectories with diverse and accurate social human motion data in a shared indoor environment

- High variety of human behavior and social interactions between individuals, groups of people and a mobile robot in presence of static obstacles
- Over 60 minutes of continuous motion tracking
- Over 600 individual and group trajectories between multiple goal points
- Accurate data for position and 3D head orientation, recorded with a motion capture system at 100 Hz
- Eye gaze data at 50 Hz with the gaze-overlaid first-person video at 25 Hz
- 3D LiDAR scans from a stationary sensor
- Video recording from a static camera
- Map of static obstacles, goal coordinates, grouping information
- Several recordings with varying obstacles and motion of the robot
Date made available2020
PublisherÖrebro University
Date of data production2020

Dataset Licences

  • CC-BY-4.0

Cite this