DeepAI
Log In Sign Up

Semi-Supervised StyleGAN for Disentanglement Learning

03/06/2020
by   Weili Nie, et al.
24

Disentanglement learning is crucial for obtaining disentangled representations and controllable generation. Current disentanglement methods face several inherent limitations: difficulty with high-resolution images, primarily on learning disentangled representations, and non-identifiability due to the unsupervised setting. To alleviate these limitations, we design new architectures and loss functions based on StyleGAN (Karras et al., 2019), for semi-supervised high-resolution disentanglement learning. We create two complex high-resolution synthetic datasets for systematic testing. We investigate the impact of limited supervision and find that using only 0.25 data is sufficient for good disentanglement on both synthetic and real datasets. We propose new metrics to quantify generator controllability, and observe there may exist a crucial trade-off between disentangled representation learning and controllable generation. We also consider semantic fine-grained image editing to achieve better generalization to unseen images.

READ FULL TEXT

page 7

page 12

page 14

page 15

page 16

page 17

page 18

page 20

10/27/2020

On the Transfer of Disentangled Representations in Realistic Settings

Learning meaningful representations that disentangle the underlying stru...
06/12/2020

LaRVAE: Label Replacement VAE for Semi-Supervised Disentanglement Learning

Learning interpretable and disentangled representations is a crucial yet...
08/08/2018

Towards Learning Fine-Grained Disentangled Representations from Speech

Learning disentangled representations of high-dimensional data is curren...
11/26/2022

Progressive Disentangled Representation Learning for Fine-Grained Controllable Talking Head Synthesis

We present a novel one-shot talking head synthesis method that achieves ...
01/23/2020

Semi-DerainGAN: A New Semi-supervised Single Image Deraining Network

Removing the rain streaks from single image is still a challenging task,...
08/25/2020

AgingMapGAN (AMGAN): High-Resolution Controllable Face Aging with Spatially-Aware Conditional GANs

Existing approaches and datasets for face aging produce results skewed t...