BriefGPT.xyz
Nov, 2017
DNA-GAN：从多属性图像中学习解缠表示
DNA-GAN: Learning Disentangled Representations from Multi-Attribute Images
HTML
PDF
Taihong Xiao, Jiapeng Hong, Jinwen Ma
TL;DR
本文提出了一种名为DNA-GAN的监督学习模型，其通过将图像的不同特征或属性分解开来，达到了逼真的图像生成和特征分离的效果，并在Multi-PIE和CelebA数据集上得到实验证明其在因素分离方面是有效的。
Abstract
disentangling
factors of variation
has always been a challenging problem in
representation learning
. Existing algorithms suffer from many
→