Rapid 3D face reconstruction from a 2D image and methods using such rapid 3D face reconstruction

摘要

Creating a 3D face reconstruction model using a single 2D image and a generic facial depth map that provides depth information. In one example, the generic facial depth map is selected based on gender and ethnicity/race. In one embodiment, a set of facial features of the 2D image is mapped to create a facial-feature map, and a 2D mesh is created using the map. The same set of facial features is also mapped onto a generic facial depth map, and a 3D mesh is created therefrom. The 2D image is then warped by transposing depth information from the 3D mesh of the generic facial depth map onto the 2D mesh of the 2D image so as to create a reconstructed 3D model of the face. The reconstructed 3D model can be used, for example, to create one or more synthetic off-angle-pose images of the subject of the original 2D image.

主权项

1. A method of identifying, using a machine, a subject in an off-angle 2D image that includes the subject's face and is taken at an off-angle, wherein the subject's face has an appearance classification, the method comprising:
receiving the off-angle 2D image containing the subject to be identified; receiving a head-on 2D image containing a face of a potential match to the subject; receiving the appearance classification of the subject's face; selecting a generic appearance-classification depth map corresponding to the received appearance classification; morphing the head-on 2D image as a function of the generic appearance-classification depth map so as to generate a 3D reconstruction of the face of the potential match; creating an off-angle 2D image of the face of the potential match as a function of the 3D reconstruction so that the off-angle 2D image of the face of the potential match is at or near the off-angle of the 2D off-angle image containing the face of the subject to be identified; and comparing the off-angle 2D image containing the face of the subject to be identified and the off-angle 2D image of the face of the potential match with one another to determine whether or not a match exists.