DEEP LEARNING TRAINING METHOD FOR MASKING ERROR REGION IN STEREO BASED RECONSTRUCTION OF HUMAN FACE

摘要

A deep learning training method for masking an error region in stereo reconstruction of a face includes the steps of preparing a face scan model, constructing a CG scene for the face scan model, and performing CG rendering on the CG scene to obtain a stereo image. generating a first depth map, performing 3D reconstruction using a stereo reconstruction algorithm with the stereo image as input, and generating a second depth map, wherein the first depth map is the second depth map generating an error mask image including an error region using