A method and apparatus for three-dimensional reconstruction of a head

Abstract

The application relates to the technical field of three-dimensional reconstruction, and provides a head three-dimensional reconstruction method and device, expression parameters are extracted from a face image and corresponding voice data respectively, a first weight of the expression parameters extracted from the voice data is adjusted according to a comparison result of a collection frame rate and a rendering frame rate of the face image, a second weight of the expression parameters is adjusted according to a distance between two human body models in a virtual space, and the extracted expression parameters and the determined weight are sent to a rendering display end, the rendering display end drives a pre-stored parameterized head model to move according to the first weight and the second weight from the expression parameters, and three-dimensional reconstruction is completed. By using the expression parameters extracted from the voice data, the problem that model reconstruction fails due to the lack of the face image can be made up, and the robustness of three-dimensional reconstruction is improved; and considering the influence of sound on facial expressions, three-dimensional reconstruction is carried out in combination with the voice data, so that the reconstructed head three-dimensional model is more realistic.

Images