The paper develops a new motion capture method from a monocular sequence of 2D perspective images. Our starting point is arm motion. We first extract and track the feature points from image sequence based on watershed seqmentation and Voronoi diagram, then by rigidity constraint and motion modelling constraint we use motion analysis method to yield 3D information of feature points. Finally the obtained data is attuned to simulate motion of model. A experiment with real images are included to demonstrate the validity of the theoretic results.
CITATION STYLE
Pan, C., & Ma, S. (1999). Motion capture of arm from a monocular image sequence. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 1614, pp. 349–357). Springer Verlag. https://doi.org/10.1007/3-540-48762-x_44
Mendeley helps you to discover research relevant for your work.