A Software Architecture for RGB-D People Tracking Based on ROS Framework for a Mobile Robot

This paper describes the software architecture of a distributed multi-people tracking algorithm for mobile platforms equipped with a RGBD sensor. Our approach features an efficient point cloud depth-based clustering, an HOG-like classification to robustly initialize a person tracking and a person classifier with online learning to drive data association. We explain in details how ROS functionalities and tools play an important role in the possibility of the software to be real time, distributed and easy to configure and debug. Tests are presented on a challenging real-world indoor environment and tracking results have been evaluated with the CLEAR MOT metrics. Our algorithm proved to correctly track 96% of people with very limited ID switches and few false positives, with an average frame rate above 20 fps. We also test and discuss its applicability to robot-people following tasks and we report experiments on a public RGB-D dataset proving that our software can be distributed in order to increase the framerate and decrease the data exchange when multiple sensors are used.