This project solves the problem of mismatching between rgb camera and depth camera of Kinect camera. And we can get higher quality point cloud model than Kinect itself. We solve the problem by first using both DLT and Zhangzhengyou‘s checkerboard to calibrate the camera, and then applying the calibrated parameters to project and re-project from the image captured by RGB and depth camera.