Automatic extraction of upper human body in single images

Extraction of the human body in single, unconstrained, monocular images is a very difficult task. Localization and extraction of the body region, however, provides important and useful knowledge that can facilitate many other tasks, such as gesture recognition, pose estimation and action recognition. In this paper we present a simple appearance-based methodology that combines face detection, skin detection, image segmentation and anthropometric constraints to efficiently estimate the position and regions of hands in images. It requires no training neither explicit estimation of the human pose. Experimental results in a difficult dataset illustrate the performance of the approach.