Multi-part body segmentation based on depth maps for soft biometry analysis

• A system for RGB-Depth human body segmentation and description is presented.
• Body clusters are automatically computed and a multi-class classifier is trained.
• 3D alignment is performed within an iterative 3D shape context fitting approach.
• We show robust biometry measurements by applying orthogonal plates to body hull.
• Results on a novel data set improve segmentation accuracy in relation to RF.

This paper presents a novel method extracting biometric measures using depth sensors. Given a multi-part labeled training data, a new subject is aligned to the best model of the dataset, and soft biometrics such as lengths or circumference sizes of limbs and body are computed. The process is performed by training relevant pose clusters, defining a representative model, and fitting a 3D shape context descriptor within an iterative matching procedure. We show robust measures by applying orthogonal plates to body hull. We test our approach in a novel full-body RGB-Depth data set, showing accurate estimation of soft biometrics and better segmentation accuracy in comparison with random forest approach without requiring large training data.