A Generative Adversarial Network Method Based on Segmentation Loss

Abstract

A generation of confrontation network method based on segmentation loss, the steps are as follows: 1. Parameter initialization: set batch size m=100, hyperparameter k=1, use Xavier method to initialize parameters, determine the maximum number of iterations and the number of iterations of loss switching parameters T, let the number of iterations epoch=0; 2, training discriminator parameters: let i=1, i is a loop variable; 3, training generator parameters; epoch=epoch+1, judge whether epoch is greater than the maximum number of iterations, such as less than the maximum The number of iterations, then repeat steps 2 and 3, if satisfied, the training ends. This method enables the generator to use different forms of loss functions in different training stages, which to a certain extent makes up for the shortcomings of the GAN theory under a single loss form, making network training more stable; by introducing feature-level losses between real samples and generated samples , making the features extracted by the discriminator more robust.

Description

technical field [0001] The invention belongs to the technical field of deep learning neural network, in particular to a generation confrontation network method based on segmentation loss. Background technique [0002] Generative Adversarial Network (GAN for short) is an unsupervised deep learning framework proposed by Goodfellow in 2014. Drawing on the idea of ​​"game theory", two players are constructed: generator (generator) and discriminator (discriminator) , the former generates images through uniform noise or Gaussian random noise with input parameters of (0, 1), and the latter discriminates the input images to determine whether the input is an image from a data set or an image generated by a generator. The discriminator feeds back the result of the judgment to the generator, so that it can be optimized toward the distribution of real data. [0003] In recent years, generative adversarial networks have been widely used in image generation and semi-supervised learning. ...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the problems existing in the above-mentioned prior art, the present invention provides a method of generative adversarial network based on segmentation loss, which can avoid the phenomenon of training instability and mode collapse of conventional generative adversarial network under a single form of loss, thereby solving the problem of discrimination The problem of poor features extracted by the generator; this method can realize that the generator uses different forms of loss functions in different training periods, and makes the network training more stable by making the generator introduce feature-level losses between real samples and generated samples. The features extracted by the machine are more robust

Images