3D tracking of honey bees

3D tracking of honey bees

The study of bee flight [1] and their collective behavior [2] has been the focus of many studies that have offered us insights into how bees are able to fly and communicate with one another. In this project we would like to 3D track honey bees to study their flight dynamics and how they respond to different flow conditions. The project would involve the manufacturing of a turbulence generating system to control flow conditions and also working with a multi-camera imaging system. We would also develop a tracking algorithm capable of extracting information about bee flight such as velocity, acceleration, and angle of attack. The project would potentially further develop to consist of working with image recognition software capable of distinguishing different types of bees.

bee tracking

Figure 1: (a) Example of a raw image from one camera showing honey bees in flight. (b) Same image as in (a) but in binary form with the locations of bees identified with green circles.


[1] Short-amplitude high-frequency wing strokes determine the aerodynamics of honeybee flight, Douglas L. Altshuler, William B. Dickson, Jason T. Vance, Stephen P. Roberts, Michael H. Dickinson; Proceedings of the National Academy of Sciences, 2005, 102 (50) 18213-18218; DOI: 10.1073/pnas.0506590102

[2] Peleg, O., Peters, J.M., Salcedo, M.K. et al. Collective mechanical adaptation of honeybee swarms. Nature Physics 14, 1193–1198 (2018); https://doi.org/10.1038/s41567-018-0262-1


  • Good mathematical understanding (in particular statistics and linear algebra)
  • Python programming
  • Experience with deep learning (PyTorch or Tensorflow recommended)


Bardia Hejazi (Max Planck Institute for Dynamics and Self-Organization)

Neural Data Science Group
Institute of Computer Science
University of Goettingen