Azure Kinect is a popular low-cost markerless Motion Capture (MoCap) system, showing promising results in clinical applications. However, during concurrent validation studies with a marker-based gold standard, reflective markers produce passive infrared (IR) noise, which significantly interferes with its tracking accuracy. In this study, we collected motion data from 15 healthy participants performing upper and lower limb exercises, concurrently recorded by Azure Kinect and the Vicon system. We found that Kinect's skeletal tracking primarily relies on IR images rather than depth images. Therefore, we developed a simple yet effective algorithm to mitigate noise in IR images. Our method significantly improved Kinect's skeletal tracking reliability, reducing missed poses from 10% to negligible levels and decreasing bone length variability across frames. Additionally, joint angle measurements improved, with lower Mean Absolute Error (MAE) in Range of Motion (ROM) and higher Intraclass Correlation Coefficient (ICC) of ROM. The code developed for this study is available at https://github.com/spongebobbe/pyKinectAzureImageManipulation.
Zaccardi, S, Brahimetaj, R, Trovalusci, F, Claeys, R, Lovecchio, R, Beckwée, D, Swinnen, E & Jansen, B 2025, Insights Into Azure Kinect Skeletal Tracking: A Simple Approach to Reduce IR Passive Noise. in 2025 25th International Conference on Digital Signal Processing (DSP). International Conference on Digital Signal Processing, DSP, pp. 1-5. https://doi.org/10.1109/DSP65409.2025.11075149
Zaccardi, S., Brahimetaj, R., Trovalusci, F., Claeys, R., Lovecchio, R., Beckwée, D., Swinnen, E., & Jansen, B. (2025). Insights Into Azure Kinect Skeletal Tracking: A Simple Approach to Reduce IR Passive Noise. In 2025 25th International Conference on Digital Signal Processing (DSP) (pp. 1-5). (International Conference on Digital Signal Processing, DSP). https://doi.org/10.1109/DSP65409.2025.11075149
@inproceedings{8988d5ca1aa14028beb9133c94fd90a7,
title = "Insights Into Azure Kinect Skeletal Tracking: A Simple Approach to Reduce IR Passive Noise",
abstract = "Azure Kinect is a popular low-cost markerless Motion Capture (MoCap) system, showing promising results in clinical applications. However, during concurrent validation studies with a marker-based gold standard, reflective markers produce passive infrared (IR) noise, which significantly interferes with its tracking accuracy. In this study, we collected motion data from 15 healthy participants performing upper and lower limb exercises, concurrently recorded by Azure Kinect and the Vicon system. We found that Kinect's skeletal tracking primarily relies on IR images rather than depth images. Therefore, we developed a simple yet effective algorithm to mitigate noise in IR images. Our method significantly improved Kinect's skeletal tracking reliability, reducing missed poses from 10% to negligible levels and decreasing bone length variability across frames. Additionally, joint angle measurements improved, with lower Mean Absolute Error (MAE) in Range of Motion (ROM) and higher Intraclass Correlation Coefficient (ICC) of ROM. The code developed for this study is available at https://github.com/spongebobbe/pyKinectAzureImageManipulation.",
author = "Silvia Zaccardi and Redona Brahimetaj and Federico Trovalusci and Reinhard Claeys and Rossana Lovecchio and David Beckw{\'e}e and Eva Swinnen and Bart Jansen",
note = "Publisher Copyright: {\textcopyright} 2025 IEEE.",
year = "2025",
month = jul,
day = "15",
doi = "10.1109/DSP65409.2025.11075149",
language = "English",
series = "International Conference on Digital Signal Processing, DSP",
pages = "1--5",
booktitle = "2025 25th International Conference on Digital Signal Processing (DSP)",
}