StyleAutoEncoder for manipulating image attributes using pre-trained StyleGAN

TL;DR

StyleAutoEncoder (StyleAE) is a lightweight plugin for pre-trained StyleGAN that enables efficient image attribute manipulation without extensive retraining, offering a cost-effective and flexible alternative to existing methods.

Contribution

Introduces StyleAutoEncoder, a simple and fast autoencoder module that enhances pre-trained StyleGAN for attribute editing, reducing computational costs and complexity.

Findings

StyleAE effectively manipulates image attributes comparable to state-of-the-art methods.

It is simpler and faster than invertible normalizing flow algorithms.

StyleAE provides greater flexibility in neural network design.

Abstract

Deep conditional generative models are excellent tools for creating high-quality images and editing their attributes. However, training modern generative models from scratch is very expensive and requires large computational resources. In this paper, we introduce StyleAutoEncoder (StyleAE), a lightweight AutoEncoder module, which works as a plugin for pre-trained generative models and allows for manipulating the requested attributes of images. The proposed method offers a cost-effective solution for training deep generative models with limited computational resources, making it a promising technique for a wide range of applications. We evaluate StyleAutoEncoder by combining it with StyleGAN, which is currently one of the top generative models. Our experiments demonstrate that StyleAutoEncoder is at least as effective in manipulating image attributes as the state-of-the-art algorithms based on invertible normalizing flows. However, it is simpler, faster, and gives more freedom in designing neural