Project information
- Category: Innovation, Software, algorithm, 3D imaging
- Project date: 2015
Automated body reconstruction and measurement using 3D imaging
We were approached for a creative concept to accurately measure body size of children and infants (Anthropometry) in the developing world as a means to assess their nutritional status. Anthropometry involves the systematic measurement of the physical properties of the human body, primarily dimensional descriptors of body size and shape. The system had to be portable such that field personnel would not need to carry high precision components to remote regions of the world. Our approach was to utilize the power of smart phones both in terms of 3D Imaging and computation power to develop a truly portable and field deployable solution. We collected data from multiple depth sensors processed them into XYZ clouds and merged them to form a single body model using sophisticated algorithms to correct for sensor aliment and calibration, pose correction, removal of outliers, and finally merging of multiple point clouds with minimal error. Once an accurate body model is available automated measurements can be taken with high reproducibility. Furthermore the model can now be processed further to identify major segment (head, torso, legs, arms), this provides additional opportunities for anthropometric studies.