The data of locations and 3-D shapes of real images in MR environment are got as depth maps measured by the SRF. The depth maps are preserved in a PC which controls the SRF. Then the data of the depth maps are transferred from the PC to a graphic workstation through Ethernet. The depth maps are transformed to polygon of 3-D models in the graphic workstation[2]. We use OpenGL for rendering virtual images and the polygon of 3-D models transformed from real images.