AUTOMATIC LANDMARKS PREDICTION USING THE ARTIFICIAL NEURAL-NETWORK-BASED TECHNIQUE ON 3D ANTHROPOMETRIC DANA
The surface of the human body offers a multitude of topological and geometric information which constitutes the main parameters to be taken into account in product and workspace design. The added components of shape provided by 3D scanning offer a more detailed description of human variation compared with traditional manual 1D or 2D data. Landmarks extracted from 3D scanning data can be considered as a reduced 3D configuration of the human body. The Automatic Landmarks Prediction Method is based on the configuration space theory which is widely applied in biology. With the configuration space theory, this paper researches Landmarks using a neural-network-based technique on 3D anthropometric data.