Instead of using 2D images to determine depth, the system uses an Asus Xtion PRO sensor to provide depth data using structured light. The data are streamed at a frame rate of 30 fps. By calibrating the depth sensor and the thermal imaging camera, as well as the intrinsic parameters of the camera, the system projects thermal images onto a mesh to generate textured coordinates. On the next slide, see how the Kinect Fusion generates the mesh.