The ability of detecting human postures is very relevant for applications related to the analysis of human behaviors. Techniques for posture detection and classification can be thus very relevant in several fields, like ambient intelligence, surveillance, elderly care, human-machine interaction. This problem has been studied in recent years in the Computer Vision community, but proposed solutions still suffer from some limitations that are due to the difficulty of dealing with complex scenes (e.g., occlusions, different view points, etc.).In this article we present a system for posture tracking and classification based on a stereo vision sensor that provides both a robust way to segment and track people in the scene and 3D information about tracked people. The proposed method is based on matching 3D data with a 3D human body model. Relevant points in the model are then tracked over time with temporal filters and a classification method based on Hidden Markov Models is used to recognize principal postures. Experimental results show the effectiveness of the system in determining human postures with different orientations of the people with respect to the stereo sensor, in presence of partial occlusions and under different environmental conditions.