Image generation method and system based on a capsule network

Abstract

The invention provides an image generation method and system based on a capsule network. The method comprises the following steps: firstly, constructing training data according to attribute characteristics of an image to be generated; Constructing an image generation model; Training the constructed image generation model by adopting a batch training method according to the training data to obtaina trained image generation model, and constructing a random noise vector; And finally, taking the attribute vector and the random noise vector of the to-be-generated image as the input of the trainedimage generation model, and generating a new image with the same size as the image in the training data. The method has the beneficial effects that the capsule network is added into the image generation model to serve as the encoder network, so that the faster convergence of the model training process is facilitated; And on the other hand, compared with the pooling process of the convolutional neural network, the dynamic routing algorithm of the capsule network has stronger generalization robustness on characteristics, and more diversified and real images can be generated.

Description

technical field [0001] The present invention relates to the field of image processing, in particular to an image generation method and system based on a capsule network. Background technique [0002] GAN models usually consist of two parts: (1) a generator, which tries to transform samples drawn from a prior distribution into samples drawn from a multi-dimensional complex data distribution; (2) a discriminator, which Decides whether the given samples are real or come from the generator's distribution. The training process of these two parts is like two players playing a game. [0003] The deep convolutional generation confrontation network DCGAN model is an extension of GAN. The convolutional network is introduced into the generative model for unsupervised training, and the powerful feature extraction ability of the convolutional network (CNN) is used to improve the learning of the generative network. Effect. DCGAN can generate high-quality images more effectively and qui...

AI Technical Summary

Problems solved by technology

[0008] In order to solve the problem that the traditional variational autoencoder VAE generates images that are blurry, unclear, and unreal, the present invention provide

Images