In this thesis, we present a user-friendly and practical method for seamless integration of computer-generated images (CG) with real photographs and video. In general such seamless integration is extremely hard and requires recovery of real world information to simulate the same environment for both CG and real objects. This real world information includes camera positions and parameters, and shapes, material properties, and motion of real objects. Among these one of the most important real world information is lighting. Image based lighting that is developed to recover illumination information of the real world from photographs has recently been popular in computer graphics. In this thesis we present a practical image based lighting method. Our method is based on a simple and easily constructible device: a square plate with a cylindrical stick. We have developed a user-guided system to approximately recover illumination information (i.e. orientations, colors, and intensities of light sources) from a photograph of this device. Our approach also helps to recover surface colors of real objects based on reconstructed lighting information. In addition, we will address our observation on shadows with multi-color lights in a compositing aspect and will present a solution to handle the addressed issue.
