Science & Engineering Research Support soCiety (SERSC)
In this paper, the authors present a novel data-driven framework for 3D full body human motion reconstruction from uncalibrated monocular video data. To this end, they develop a knowledge base by taking 2D samples of the motion capture library from different viewing directions. This allows later steps to handle 2D query videos without any information on the viewing direction. They detect and track features from input video sequences by utilizing low-level image based feature detection techniques like MSER (Maximally Stable Extremal Region) and SURF (Speeded Up Robust Feature).