Online Database of Gestures in hdf5 Format


Remark  This data was recorded in public funded projects and corporations with Daimler AG. We will not answer any question about the recorded data. All information is included in the following lists.

Dataset Description

Scenario List

  • Signalling to slow down with two hands (Radar at front)
  • Walking (Radar at front)
  • Running (Radar at front)
  • Walking and then falling down (Radar at front)
  • Waving with both hands (Radar at front)
  • Walking with crutches (Radar at front)


  • Each scenario has measurements from 13 persons and from each person there are 10 measurements
  • The persons are not corresponding to each other, in other words, person 1 in scenario1 does not have to be the same person as in scenario2

Measurement Campaign, hdf5 Files

  • All the measurements are recorded as hdf5 and contain all the relevant information about the measurement with all the measurement parameters and measurement data
  • The hdf5 file contains the parameters information in the attributes field.
    Attribute Number Description
    N 512 number of samples in one ramp
    fStrt 76 GHz start frequency
    fstop 77 GHz stop frequency
    TRampDo 0.000005 Time of the down ramp in seconds
    TRampUp 0.000051 Time of the up ramp in seconds
    NrFrms 256 Number of frames
    NLoop 512 Number of ramps in a frame
  • And the IF-signal information in the dataset “datachn” field.
  • Notice that the chirp counter is included in the first cell.

If you use this database, please use these references in your reference list

  • K. Ishak, N. Appenrodt, J. Dickmann and C. Waldschmidt, "Human Motion Training Data Generation for Radar Based Deep Learning Applications," IEEE MTT-S International Conference on Microwaves for Intelligent Mobility (ICMIM), Munich, 2018, pp. 1-4, doi: 10.1109/ICMIM.2018.8443559.
  • K. Ishak, N. Appenrodt, J. Dickmann and C. Waldschmidt, "Advanced Radar Micro-Doppler Simulation Environment for Human Motion Applications," IEEE Radar Conference (RadarConf), Boston, MA, USA, 2019, pp. 1-6, doi: 10.1109/RADAR.2019.8835755.