Styleformer: Transformer based Generative Adversarial Networks with Style Vector
Preprint
Abstract
We propose Styleformer, which is a style-based generator for GAN architecture, but a convolution-free transformer-based generator. In our paper, we explain how a transformer can generate high-quality images, overcoming the disadvantage that convolution operations are difficult to capture global features in an image. Furthermore, we change the demodulation of StyleGAN2 and modify the existing transformer structure (e.g., residual connection, layer normalization) to create a strong style-based generator with a convolution-free structure. We also make Styleformer lighter by applying Linformer, enabling Styleformer to generate higher resolution images and result in improvements in terms of speed and memory. We experiment with the low-resolution image dataset such as CIFAR-10, as well as the high-resolution image dataset like LSUN-church. Styleformer records FID 2.82 and IS 9.94 on CIFAR-10, a benchmark dataset, which is comparable performance to the current state-of-the-art and outperforms all GAN-based generative models, including StyleGAN2-ADA with fewer parameters on the unconditional setting. We also both achieve new state-of-the-art with FID 15.17, IS 11.01, and FID 3.66, respectively on STL-10 and CelebA. We release our code at https://github.com/Jeeseung-Park/Styleformer.Keywords
All Related Versions
This publication has 21 references indexed in Scilit:
- Big Transfer (BiT): General Visual Representation Learning2020
- Analyzing and Improving the Image Quality of StyleGAN2020
- AdversarialNAS: Adversarial Neural Architecture Search for GANs2020
- Escaping from Collapsing Modes in a Constrained Space2018
- StarGAN: Unified Generative Adversarial Networks for Multi-domain Image-to-Image Translation2018
- Image Generation from Scene Graphs2018
- Generative Image Inpainting with Contextual Attention2018
- CLEVR: A Diagnostic Dataset for Compositional Language and Elementary Visual Reasoning2017
- Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network2017
- The Cityscapes Dataset for Semantic Urban Scene Understanding2016