Artificial neural networks in the transformation of human body posture

Designers, engineers and ergonomists are seeking to
exploit the opportunities offered by the 3D
anthropometric technologies. These technologies make
3D measurements possible and provide us with a more
detailed description of human body in comparison with
the traditional 1D or 2D data processing. In many
industrial design cases, there is a need to take
into consideration various postures of the human
body when the product is designed. This book presents
an approach to transforming measured body data
between various postures. In this research the
measured human body data were substituted by a proper
set of landmarks which was used as a basis of
transforming the specific body postures. Artificial
neural networks have been used for the actual
conversion of data. The input consisted of a set of
demographic data and the set of coordinates of the
landmarks characterizing a given posture. The output
was another set of landmarks describing the
transformed posture. The results have showed that the
ANNs and landmark-based posture prediction technology
is computationally effective. This technology is
generally applicable and opens up new possibilities
in studying motions and hand postures.

Autorentext
B.Zhang worked as lecturer and product designer at the Hunan University,P.R.China; studied as Ph.D.candidate in Applied Ergonomics of the Faculty of Industrial Design Engineering, TU Delft. J.F.M.Molenbroek, Associate Professor Applied Ergonomics at Delft University of Technology, co-editor at The Ergonomics Open Journal; Chair at Ergindo.

Klappentext
Designers, engineers and ergonomists are seeking to exploit the opportunities offered by the 3D anthropometric technologies. These technologies make 3D measurements possible and provide us with a more detailed description of human body in comparison with the traditional 1D or 2D data processing. In many industrial design cases, there is a need to take into consideration various postures of the human body when the product is designed. This book presents an approach to transforming measured body data between various postures. In this research the measured human body data were substituted by a proper set of landmarks which was used as a basis of transforming the specific body postures. Artificial neural networks have been used for the actual conversion of data. The input consisted of a set of demographic data and the set of coordinates of the landmarks characterizing a given posture. The output was another set of landmarks describing the transformed posture. The results have showed that the ANNs and landmark-based posture prediction technology is computationally effective. This technology is generally applicable and opens up new possibilities in studying motions and hand postures.