Method for image processing based on matrix variable variation self-encoder

Abstract

The invention discloses a method for image processing based on a matrix variable variation self-encoder so that a problem of damaging the spatial structure by image vectorization processing can be solved and thus image reconstruction, denoising and completion can be realized. Different from the traditional VAE, the method has the following advantages: the input, hidden layer features, latent variable distribution parameters of the model are described by using an inherent expression form -2D matrix of the image; with the definition of matrix Gaussian distribution and related properties, an explicit expression of an objective function of a novel model is deduced and then model parameters are calculated by using a stochastic gradient descent algorithm. In the model, modeling of the spatial structure and statistical information of the image data is carried out well because the involved modeling process is oriented at the matrix variable, so that the image reconstruction quality is enhanced, noises are removed, and the image completion is realized.

Description

technical field [0001] The invention belongs to the field of computer vision and machine learning, and in particular relates to an image processing method based on a matrix variable variational autoencoder. Background technique [0002] Image reconstruction, denoising and completion are important contents of image processing. Variational Autoencoder (VAE) is widely used in image processing related fields because it can well model the probability distribution of image data. VAE usually consists of an inference model (encoder) and a generation model (decoder). The objective function of the model includes two items: one is the reconstruction error of the image, which is generally measured by mean square error or cross entropy; the other is The Kullback-Leibler (KL) divergence is used to measure the similarity between the posterior of the feature latent variables learned based on the inference model and the prior distribution of the feature assumptions, which is equivalent to a ...

AI Technical Summary

Problems solved by technology

[0004] The technical problem to be solved by the present invention is to provide a method for image processing based on a matrix variable variational autoencoder (Matrix-variate Variational Autoencoder, MVVAE), which can solve the problem of image vectorization processing destroying the spatial structure, thereby benefiting the image Refactoring, denoising and completion

Images