A method and device for determining a face model based on artificial intelligence

Abstract

The embodiment of the invention discloses a face model determination method based on artificial intelligence. The method comprises the steps of obtaining collection data for a to-be-built object, including multiple groups of face vertex data, identifying a face image of the to-be-built object in each group, and determining initial face pinching parameters based on first vertex data in the face images and target vertex data used for identifying a target face model; obtaining second vertex data in collected data, determining expression parameters based on initial face pinching parameters and target vertex data, then fixing the expression parameters, and determining target face pinching parameters according to the second vertex data and the target vertex data. The second vertex data and the first vertex data respectively identify different face images; the probability that the first vertex data and the second vertex data are influenced by identical uncertainty is low, the noise in the first vertex data is offset to a certain extent through the target face pinching parameter corrected through the second vertex data, and the accuracy of the face model corresponding to the to-be-built object determined according to the target face pinching parameter is relatively higher.

Description

technical field [0001] The present application relates to the field of data processing, in particular to a method and device for determining a face model based on artificial intelligence. Background technique [0002] At present, human-computer interaction is relatively common, and users can interact with animated images. For any segment of speech, an animated image can be driven to make the mouth shape corresponding to the segment of speech. In this scenario, the presence of animated images can greatly enhance the sense of reality, improve expressiveness, and bring users a more immersive experience. [0003] The animated image can be a virtual image such as a cartoon or a virtual person, or can be a simulated image constructed based on the appearance of a real person. When constructing an animation image through the object to be built, the object to be built can be video collected, and the face model of the object to be built can be determined through the collected data. ...

AI Technical Summary

Problems solved by technology

[0004] However, in related technologies, simple devices such as mobile phones are often used to collect video of the object to be built, and the collection environment and method are generally relatively random. As a result, in such high-uncertainty scenarios, the collected data is often low in accuracy and noisy. Larger, the quality of the face model established based on the collected data is not high, and it is difficult to accurately reflect the actual shape of the object to be built

Images