Image generation method based on Gaussian mixture model prior variation auto-encoder

Abstract

The invention discloses an image generation method based on a Gaussian mixture model prior variation auto-encoder. The method comprises the following steps: S11, presetting and generating an image training data set, wherein the training data set is composed of a plurality of batches of training data; s12, building a variation auto-encoder network of Gaussian mixture model prior; s13, uploading a plurality of preset batches of training data to the variational auto-encoder network, and determining posteriori distribution and prior distribution of the variational auto-encoder network; s14, determining a relationship between Gaussian components in the Gaussian mixture model to obtain a mapping function; s15, obtaining a reconstruction loss function and a KL divergence function by using the variational auto-encoder network and the obtained mapping function, calculating a posteriori distribution loss function and a priori distribution loss function of the variational auto-encoder network, and updating parameters of the variational auto-encoder network to generate an image; and S16, when an image is generated, taking the pseudo input as an input image and uploading the pseudo input to thevariational auto-encoder network to obtain a finally generated picture.

Description

technical field [0001] The invention relates to the technical field of deep learning, in particular to an image generation method based on a Gaussian mixture model prior variational autoencoder. Background technique [0002] In the Internet age, machine learning has developed rapidly and made great achievements. Among them, image generation technology, as a branch of machine learning, has played an important role in our understanding of images. The image generation model is a probability model for probabilistic modeling of images, and the deep neural network can be regarded as a very complex nonlinear function with strong fitting ability, which can be used to build a generation model to estimate the probability density function parameters. The image generation model can be used to generate more different image samples, to restore image information, to convert images of different modalities or between images, text, and voice, and to predict the future. For example, future f...

AI Technical Summary

Problems solved by technology

However, standard variational autoencoder priors tend to produce blurry images due to underfitting

Images