Real-time camera tracking is steadily gaining in importance due to the drive from various applications, such as AR(augmented reality), mobile computing, and human-machine interface. In this paper, we describe a real-time camera tracking framework designed to track a monocular camera in a desktop workspace. Basic idea of the proposed scheme is that the camera pose estimation is achieved on the basis of a planar object tracking framework. As the camera pose estimation and scene registration is achieved via a non-iterative process, the proposed method is computationally efficient and very fast, and therefore, it can be directly embedded to AR systems running on mobile device platforms. In addition, our system attempts to detect new features assumed to be present on the reference planar surface, so that the system can be operated even when reference features go out of visible range. The accuracy and robustness of the proposed system are verified on the experimental results of several real-time input video streams.