Automatic posing of a meshed human model using point clouds

Multilevel active registration for kinect human body scans: from low quality to high quality

“…If two shapes are perfectly isometric, then there exists an isometry i.e., a distance-preserving mapping, between these shapes such that the geodesic distance between any two points on one shape is exactly the same as the geodesic distance between their correspondences on the other [36]. Different approaches are proposed to exploit isometry for shape correspondences [14,15,20,29,35]. One way is to embed shape into a different domain where geodesic distances are replaced by Euclidean distance so that isometric deviation can be measured and optimized in the embedding space [15].…”

“…One way is to embed shape into a different domain where geodesic distances are replaced by Euclidean distance so that isometric deviation can be measured and optimized in the embedding space [15]. Euclidean embedding can be achieved using various techniques such as classical MDS (Multidimensional Scaling) [20,35], least-squares MDS [15], and spectral analysis of the graph Laplacian [29] or of the Laplace-Beltrami operator [14]. However, when it comes to the meshes from low-cost scanners, the above isometry-based methods are not applicable as they usually require watertight meshes and suffer from self-symmetry of human body shape.…”

“…To register the 3D scans, several 3D fitting methods are proposed [1,2,5,14,32,49]. The invertible finite volume method [14] is used to control the template tetrahedral mesh to the target point clouds. The stitched puppet model [49] adopts the DPMP algorithm which is a particle-based method to align a graphical model to target meshes.…”

“…Current publicly available high-quality human body datasets, such as SCAPE [3], FAUST [5], TOSCA [8] are built either from costly laser scanners or need other assistance (e.g makers, texture or professional tools). With the appearance of low-cost scanners such as Kinect, it is now possible for an object, a room or even a person to be quickly scanned, modeled and tracked [12,14,27,28,30,31,44]. Nowadays human body meshes could be captured for different identities in different poses in a few minutes.…”

“…To register the 3D scans, several 3D fitting methods are proposed [1,2,5,14,32,49]. The invertible finite volume method [14] is used to control the template tetrahedral mesh to the target point clouds.…”