Research on face tracking method based on improved CAMSHIFT and Kalman filter

With the rapid development of information technology, biometric recognition technologies have become a research hotspot in the field of pattern recognition due to their security, reliability, and effectiveness. Face detection and tracking technologies have become an active exploration topic for researchers both domestically and internationally, as they combine intelligence with everyday life. This paper employs a skin color model-based method for face detection, while for face tracking, an improved combination of the Kalman filter algorithm and the Camshift algorithm is used. The aim is to construct an efficient model system for face tracking. The movement or rapid changes in the face do not interrupt the tracking process, which enhances the real-time performance and robustness of face tracking.