Authors: Jaeyoung Yoo,Hojun Lee,Nojun Kwak
ArXiv: 1902.04294
Document:
PDF
DOI
Abstract URL: http://arxiv.org/abs/1902.04294v1
In this paper, we treat the image generation task using the autoencoder, a
representative latent model. Unlike many studies regularizing the latent
variable's distribution by assuming a manually specified prior, we approach the
image generation task using an autoencoder by directly estimating the latent
distribution. To do this, we introduce 'latent density estimator' which
captures latent distribution explicitly and propose its structure. In addition,
we propose an incremental learning strategy of latent variables so that the
autoencoder learns important features of data by using the structural
characteristics of under-complete autoencoder without an explicit
regularization term in the objective function. Through experiments, we show the
effectiveness of the proposed latent density estimator and the incremental
learning strategy of latent variables. We also show that our generative model
generates images with improved visual quality compared to previous generative
models based on autoencoders.