Learning Inverse Mapping by Autoencoder based Generative Adversarial Nets

TL;DR

This paper introduces a novel autoencoder-based method to learn the inverse of GAN generators, improving training stability and performance in image searching and translation tasks.

Contribution

It proposes using an inverse generator as encoder and a pre-trained generator as decoder within an autoencoder framework to effectively learn the inverse mapping of GANs.

Findings

Outperforms traditional inverse learning methods in image searching

Enhances training stability for inverse GAN models

Demonstrates effectiveness in image translation tasks

Abstract

The inverse mapping of GANs'(Generative Adversarial Nets) generator has a great potential value.Hence, some works have been developed to construct the inverse function of generator by directly learning or adversarial learning.While the results are encouraging, the problem is highly challenging and the existing ways of training inverse models of GANs have many disadvantages, such as hard to train or poor performance.Due to these reasons, we propose a new approach based on using inverse generator ($IG$) model as encoder and pre-trained generator ($G$) as decoder of an AutoEncoder network to train the $IG$ model. In the proposed model, the difference between the input and output, which are both the generated image of pre-trained GAN's generator, of AutoEncoder is directly minimized. The optimizing method can overcome the difficulty in training and inverse model of an non one-to-one function.We also applied the inverse model of GANs' generators to image searching and translation.The experimental results prove that the proposed approach works better than the traditional approaches in image searching.