Method and system for three-dimensional face reconstruction based on binocular stereo and photometric stereo

Abstract

The invention relates to the technical field of computer vision. The embodiment specifically discloses a method and a system for three-dimensional reconstruction of a human face based on binocular stereo and photometric stereo. The initial depth image of a target human face is obtained by binocular stereo matching. Four images of the target face collected by two light sources and two camera devices are represented by photometric stereo as constraint terms and combined with the initial depth image to construct the energy function of the depth image to be solved. The optimized depth image is obtained by minimizing the energy function, and the optimized depth image is transformed into a three-dimensional point cloud to obtain the three-dimensional point cloud of the target face. It solves theproblem that the existing photometric stereo vision needs at least three light sources for 3D reconstruction, and the scale-free feature has deformation, and realizes the accurate 3D reconstruction of human face without deformation.

Description

technical field [0001] The invention relates to the technical field of computer vision, in particular to a method and system for three-dimensional face reconstruction based on binocular stereo and photometric stereo. Background technique [0002] At present, the 3D reconstruction methods of human face mainly include four kinds: parametric model, physical muscle model, multi-image visual model and 3D deformation. The face parameter model and physical muscle model are complex, computationally intensive, and lack realism; multi-image visual modeling uses multiple face images or video sequences to restore the 3D information of the face through a certain visual algorithm. Here Including monocular vision, binocular vision and multi-eye vision. Monocular vision generally obtains feature points through a single face photo, and then performs global and local fine deformation on a general-purpose 3D face mesh model; binocular vision generally uses two face images from different persp...

AI Technical Summary

Problems solved by technology

This type of algorithm is greatly affected by image quality, so the robustness is not very high, and the results are not very accurate

At present, photometric stereo vision can be used for 3D reconstruction to obtain an accurate 3D model, but at least three light sources are required, and since photometric stereo has no scale features, the reconstructed 3D model will have a certain degree of deformation

Images