In recent years, there have been more and more research and applications on three-dimensional model reconstruction. With the help of Kinect, a somatosensory device invented by Microsoft, we realized three-dimensional model reconstruction process. Firstly, this paper introduced the acquisition mechanism of color and depth images. Because the color camera and the depth camera are not coincident, precise alignment of them is presented. Then, the depth image is filtered to remove small gaps using median filtering method. For bigger black portions, the interpolation algorithm is proposed. Finally, with the tridimensional triangularization algorithm, the three-dimensional scene is reconstructed with both color and depth information. The experimental results demonstrate that the proposed algorithm can produce high quality maps.