A Perspective Approach Toward Monocular Particle-based People Tracking
Christian Waechter, Daniel Pustka, Gudrun Johanna Klinker · 2009
We present a solution to the people tracking problem using a monocular vision approach from a birds eye view and three-dimensional human representations maintained by Sequential Monte-Carlo Filtering. The system robustly estimates the image measurements likelihoods imitating the image formation process and respecting partial occlusion by dynamic object movements. Due to algorithmic optimization the system is real-time capable and can be used for Human-Computer-Interaction (HCI) in combination with other sensors. The approach can also be extended to multi-view camera systems.