Model-Based Dynamic Human Tracking and Reconstruction During Dynamic SLAM

Most of the existed Simultaneous Localization and Mapping solutions cannot work in dynamic environments since the dynamic objects lead to wrong uncertain feature associations. In this paper, we involved a learning-based object classification front end to recognize and remove the dynamic object, and thereby ensure our ego-motion estimator’s robustness in high dynamic environments. The static backgrounds are used for static environment reconstruction, the extracted dynamic human objects are used for human object tracking and reconstruction. Experimental results show that the proposed approach can provide not only accurate environment maps but also well-reconstructed moving humans.