Experimental results of 2D depth-depth matching algorithm based on depth camera Kinect v1

In the last year, we proposed a smart transcription algorithm. In the algorithm, a real liver is always captured by 3D depth camera. As contrasted with this, its virtual liver is represented by a polyhedron with STL (Standard Triangulated Language) format (Stereo-lithography) via DICOM (Digital Imaging and Communication in Medicine) data captured by MRI (Magnetic Resonance Imaging) and/or CT (Computed Tomography) scanner. By comparing a depth image in a real world and the Z-buffer in its virtual world, we quickly identify translation/rotation differences between real and virtual livers in GPU (Graphics Processing Unit). Then by a randomized steepest descent method based on the differences, we can quickly copy real lever motion to virtual liver motion. In this paper, this performance (motion precision and calculation time) of the proposed algorithm is ascertained in several kinds of experiments based on the depth camera Kinect v1. This is the first challenge to use our smart algorithm running in a 3D real environment.