Markerless Motion Capture from Monocular Videos

We present a method to determine the 3D spatial locations of joints of a human body from a monocular video sequence of a Bharatanatyam dance. The proposed method uses domain specific knowledge to track major joints of the human in motion from the two dimensional input data. We then make use of various physical and motion constraints regarding the human body to construct a set of feasible 3D poses. A dynamic programming based method is used to find an optimal sequence of feasible poses that represents the original motion in the video.

• Type: Conference
• Year: 2004
• Where: ICVGIP
• Authors: Vishal Mamania, Appu Shaji, Sharat Chandran
• Keywords: Feasible Poses, Human Body, ICVGIP 2004, ICVGIP 2008, Monocular Video Sequence