Style digital human generation method and device, equipment and readable storage medium

Abstract

The present application relates to the technical field of artificial intelligence, and more particularly to a style digital person generation method, device and equipment and readable storage medium, which introduces the stylization of images into the education industry of financial knowledge popularization, and the method comprises the following steps: obtaining input text, inputting the text into a text editor to obtain a real person image; inputting the real person image into a generator to generate a human body surface graph; performing UV unfolding on the human body surface graph to obtain a UV map of a visible area; and inputting the UV map of the visible area into a trained neural network to obtain a stylized digital person. The style digital person generation method can effectively separate the human body surface graph from UV unfolding, effectively separate the person and the background, set the virtualization of the person, adjust the style according to the corresponding style, and set different parameters according to the background, so that different setting styles are fully generated, and the stylized digital person is obtained according to the trained neural network.

Images