A 3D Animation Generation Method Based on Deep Recurrent Neural Network Algorithm

Abstract

A 3D animation generation method based on a deep recurrent neural network algorithm. With reference to Figure 1, the technical route to be adopted by this method is given, and the existing 3D animation data set is used as a drive to extract the dynamic features of 3D animation. Above all, the technical route of generating 3D animation by integrating deep neural network model technology and carrying out related theories and algorithm design, the main beneficial effect of the present invention is to propose a data-driven 3D animation generation method, based on real 3D animation data, and integrating deep neural network technology, Extract the dynamic features of the training data and load them into a given grid model to realize animation generation.

Description

Technical field[0001]The present invention relates to the field of computers, in particular to a method for generating three-dimensional animation based on a deep loop neural network algorithm.Background technique[0002]3D Animation Generation: By inputting a 3D model, it automatically extracts the behavior of the specified 3D animation and drives the given 3D model, thereby driving the given model to generate 3D animations with similar behaviors.[0003]Artificial Neural Network (ANN): The artificial neural network model includes an input layer, a hidden layer, and an output layer. It simulates the principles of the nervous system in the human brain and is achieved by adjusting the interconnection between the nodes in the neural network model. Information processing. The neural network model can theoretically approach any objective function infinitely.[0004]Deep learning (Deep Learning, DL): Deep learning is a new research field in machine learning research. It is an extension of arti...

AI Technical Summary

Problems solved by technology

On the other hand, the application of the existing deep neural network model to the generation of 3D animation still faces some problems: 1) the descriptors describing the dynamic characteristics of 3D animation are limited; 3) There is no unified method or criterion for evaluating the generation effect of 3D animation.

[0014] Based on relevant research at home and abroad, we found that so far, research based on deep learning algorithms has mainly focused on the network structure design of deep neural network models, as well as the application of pattern recognition and data generation of data including images, texts, and dialogues. , but the discussion and research on the application of behavior analysis and data generation for 3D animation is seriously insufficient

Images