Azure Kinect 3D skeleton and foot pressure data for pathological gaits

Citation Author(s):
Kooksung
Jun
Gwangju Institute of Science and Technology, school of integrated technology
Sanghyub
Lee
Gwangju Institute of Science and Technology, school of integrated technology
Deok-Won
Lee
Gwangju Institute of Science and Technology, school of integrated technology
Mun Sang
Kim
Gwangju Institute of Science and Technology, school of integrated technology
Submitted by:
Kooksung Jun
Last updated:
Wed, 12/01/2021 - 20:24
DOI:
10.21227/ev8a-wr16
Data Format:
License:
0
0 ratings - Please login to submit your rating.

Abstract 

Sequential skeleton and average foot pressure data for normal and five pathological gaits (i.e., antalgic, lurching, steppage, stiff-legged, and Trendelenburg) were simultaneously collected. The skeleton data were collected by using Azure Kinect (Microsoft Corp. Redmond, WA, USA). The average foot pressure data were collected by GW1100 (GHIWell, Korea). 12 healthy subjects participated in data collection. They simulated the pathological gaits under strict supervision. A total of 1,440 data instances (12 people x 6 gait types x 20 walkings) were collected.

Instructions: 

The database contains two types of pathological gait datasets.

  1. Skeleton data obtained by using Azure Kinect (size = sequences x 97)
  2. Average foot pressure data obtained by GW1100 (size = 48 x 128)

Azure Kinect Skeleton

The skeleton data are formatted as (time, joint0_x, joint0_y, joint0_z, ... , joint31_x, joint31_y, joint31_z). We captured the skeleton data of 4-m walkway. The joint numbers can be found in https://docs.microsoft.com/ko-kr/azure/kinect-dk/body-joints.

They were calibrated as below:

  • x_axis: orthogonal to walking direction
  • y_axis: walking direction
  • z_axis: normal to the ground, upward

Average Foot Pressure

GW1100 is a 1080mm x 480mm sized pressure plate and contains 6,144 high-voltage matrix sensors with maximum pressure 100 N/cm^2.

 

Reference

K. Jun, S. Lee, D. -W. Lee and M. S. Kim, "Deep learning-based multimodal abnormal gait classification using a 3D skeleton and plantar foot pressure," IEEE Access, doi: 10.1109/ACCESS.2021.3131613.

 

Related Works

  1. K. Jun, Y. Lee, S. Lee, D.-W. Lee, and M. S. Kim, "Pathological Gait Classification Using Kinect v2 and Gated Recurrent Neural Networks," IEEE Access, vol. 8, pp. 139881-139891, 2020.
  2. K. Jun, D. W. Lee, K. Lee, S. Lee, and M. S. Kim, “Feature Extraction Using an RNN Autoencoder for Skeleton-based Abnormal Gait Recognition”, IEEE Access, vol. 8, pp. 19196-19207, 2020.

  3. D. W. Lee, K. Jun, S. Lee, J. K. Ko, and M. S. Kim, “Abnormal gait recognition using 3D joint information of multiple Kinects system and RNN-LSTM,” In 2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), 2019, pp.  542-545.

Contact

kooksung930@gm.gist.ac.kr