An Articulating Statistical Shape Model of the Human Hand

Publication Type:

Conference Paper

Source:

Advances in Human Factors in Simulation and Modeling (AHFE 2018), Springer International Publishing, Volume 780, Cham, p.433–445 (2019)

ISBN:

978-3-319-94223-0

Abstract:

This paper presents a registration framework for the construction of a statistical shape model of the human hand in a standard pose. It brings a skeletonized reference model of an individual human hand into correspondence with optical 3D surface scans of hands by sequentially applying articulation-based registration and elastic surface registration. Registered surfaces are then fed into a statistical shape modelling algorithm based on principal component analysis. The model-building technique has been evaluated on a dataset of optical scans from 100 healthy individuals, acquired with a 3dMD scanning system. It is shown that our registration framework provides accurate geometric and anatomical alignment, and that the shape basis of the resulting statistical model provides a compact representation of the specific population. The model also provides insight into the anatomical variation of the lower arm and hand, which is useful information for the design of well-fitting products.

Research area: