Method for automatically generating virtual human animation based on text

Abstract

The invention relates to a method for automatically generating a virtual human animation based on a text, and the method comprises the following steps: carrying out the preprocessing of an input text, screening out non-null characters, and carrying out the sentence segmentation of the text; for the clauses, using a word segmentation model and corpus data to perform word information analysis; using a specific behavior emotion model and rule to identify the behavior emotion contained in the sentence, and outputting a behavior emotion label; screening and filtering the hit behavior emotion tags according to preset rule logic; mapping the filtered behavior emotion to a preset animation segment through a mapping relation; joining and rendering a plurality of animation segments corresponding to the text; and outputting the final virtual human animation video. According to the method, the defects in the prior art are overcome, the natural language understanding technology and the computer animation technology are combined, the virtual human is supported to automatically generate corresponding behavior emotion expressions based on any text verbal skill, and the virtual human is made to have intelligent body languages.

Description

technical field [0001] The invention relates to the technical fields of artificial intelligence and computer animation, in particular to a method for automatically generating virtual human animation based on text. Background technique [0002] Virtual human animation technology has been widely used in education and training, advertising marketing, film and television animation and many other fields. The traditional production method is that professional planners formulate storyboard scripts according to specific storylines and character speech skills, and then professional artists Use tool software to make animations of the limbs and expressions of virtual characters, and finally output them as animated videos by means of graphics rendering and post-processing. The advantage of the traditional method is that the behavior and expression of the avatar can be designed and produced according to the specific application scenarios, and there is no need to invest a lot of energy in...

AI Technical Summary

Problems solved by technology

[0004] 1) The virtual human does not have the ability to automatically analyze behavioral emotions based on the semantic information contained in the text, and finally automatically take corresponding actions

[0005] 2) In previous studies in the field of artificial intelligence, the classification of emotions is usually coarse-grained, which cannot provide sufficient support for the rich emotional animation performance of virtual humans

[0006] 3) In the previous research in the field of artificial intelligence, there were few analyzes specifically aimed at behavior and gestures, which could not provide sufficient support for the rich body and gesture animation performance of virtual humans

Images