Automatic Generation Technology of Animated Character Expression and Action Based on Deep Learning

Based on deep learning technology, this paper focuses on the automatic generation technology of animated characters' expressions and actions. By extensively collecting the expression and action data of animated characters and constructing rich data sets, convolutional neural network and recurrent neural network models are used for training. The expression generation model is designed, and its accuracy, realism and consistency in expression generation are evaluated by performance index. In the experiment, the learning rate, the number of layers in the network and the size of the convolution kernel are adjusted, and the influence of these factors on the generation effect is discussed. It provides a new method for the animation industry to accelerate the production process, improve the realism of expression and action, and expand the artistic quality and viewing experience of animation works.