摘要 |
<p>Disclosed is a method and apparatus for generating a three-dimensional (3D) face image. The method for generating a 3D face image may include a step of generating a second 3D face model with a first 3D face model based on a feature point of a 2D face and a step of matching the boundary of the second 3D face model and the boundary of the 2D face. The feature point of the 2D face is a point including information on a structural feature of the 2D face, and the second 3D face model could be a final fitted model by reflecting information on the structural feature of the 2D face to the first 3D face model. Therefore, the 3D face with small visual distortion may be generated by accurately predicting depth information on a detailed part of the 2D face.</p> |