Multi-View Skeleton Pose Sequence Selection for Human Activity Analysis

Vision-based pose estimation and tracking of a human body is a problem of estimating
kinematic parameters of the body model, such as joints position and joints angle, from a static
image or a video sequence. In this paper, an estimation of a human pose using a multi-view
camera system is performed. The constructed system is using a few stationary calibrated
Microsoft Kinect devices, giving the posesequence estimation for each of the views.

• Petar Nikolov
• Ognian Boumbarov

International Journal of Innovative Science, Engineering & Technology, vol. 4, no. 7.

https://ijiset.com/vol4/v4s7/IJISET_V4_I07_04.pdf