# TransGAN: Two Pure Transformers Can Make One Strong GAN, and That Can Scale Up

arXiv: https://arxiv.org/abs/2102.07074 · HTML: https://mlsanity.com/paper/2102.07074/

## Key facts

- Cited by 162 papers in the map (data through 14 Oct 2026).
- Builds on 68 of its 87 references that are in the map.
- First posted on arXiv 14 Feb 2021.

## Builds on (most cited first, 30 of 68)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [End-to-End Object Detection with Transformers](https://mlsanity.com/paper/2005.12872/) (2020)
- [Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift](https://mlsanity.com/paper/1502.03167/) (2015)
- [A Style-Based Generator Architecture for Generative Adversarial Networks](https://mlsanity.com/paper/1812.04948/) (2018)
- [Layer Normalization](https://mlsanity.com/paper/1607.06450/) (2016)
- [Training data-efficient image transformers & distillation through attention](https://mlsanity.com/paper/2012.12877/) (2020)
- [Image-to-Image Translation with Conditional Adversarial Networks](https://mlsanity.com/paper/1611.07004/) (2016)
- [Deep Learning Face Attributes in the Wild](https://mlsanity.com/paper/1411.7766/) (2014)
- [Analyzing and Improving the Image Quality of StyleGAN](https://mlsanity.com/paper/1912.04958/) (2019)
- [Deformable DETR: Deformable Transformers for End-to-End Object Detection](https://mlsanity.com/paper/2010.04159/) (2020)
- [Longformer: The Long-Document Transformer](https://mlsanity.com/paper/2004.05150/) (2020)
- [Taming Transformers for High-Resolution Image Synthesis](https://mlsanity.com/paper/2012.09841/) (2020)
- [Progressive Growing of GANs for Improved Quality, Stability, and Variation](https://mlsanity.com/paper/1710.10196/) (2017)
- [Improved Techniques for Training GANs](https://mlsanity.com/paper/1606.03498/) (2016)
- [Large Scale GAN Training for High Fidelity Natural Image Synthesis](https://mlsanity.com/paper/1809.11096/) (2018)
- [Non-local Neural Networks](https://mlsanity.com/paper/1711.07971/) (2017)
- [Improved Training of Wasserstein GANs](https://mlsanity.com/paper/1704.00028/) (2017)
- [Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks](https://mlsanity.com/paper/1511.06434/) (2015)
- [Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network](https://mlsanity.com/paper/1609.04802/) (2016)
- [Spectral Normalization for Generative Adversarial Networks](https://mlsanity.com/paper/1802.05957/) (2018)
- [Is Space-Time Attention All You Need for Video Understanding?](https://mlsanity.com/paper/2102.05095/) (2021)
- [Generating Long Sequences with Sparse Transformers](https://mlsanity.com/paper/1904.10509/) (2019)
- [Point Transformer](https://mlsanity.com/paper/2011.00931/) (2020)
- [Rethinking Semantic Segmentation from a Sequence-to-Sequence Perspective with Transformers](https://mlsanity.com/paper/2012.15840/) (2020)
- [LSUN: Construction of a Large-scale Image Dataset using Deep Learning with Humans in the Loop](https://mlsanity.com/paper/1506.03365/) (2015)
- [Instance Normalization: The Missing Ingredient for Fast Stylization](https://mlsanity.com/paper/1607.08022/) (2016)
- [Training Generative Adversarial Networks with Limited Data](https://mlsanity.com/paper/2006.06676/) (2020)

## Built on by (most cited first, 30 of 162)

- [Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow](https://mlsanity.com/paper/2209.03003/) (2022)
- [Consistency Models](https://mlsanity.com/paper/2303.01469/) (2023)
- [One-step Diffusion with Distribution Matching Distillation](https://mlsanity.com/paper/2311.18828/) (2023)
- [Uformer: A General U-Shaped Transformer for Image Restoration](https://mlsanity.com/paper/2106.03106/) (2021)
- [AdaptFormer: Adapting Vision Transformers for Scalable Visual Recognition](https://mlsanity.com/paper/2205.13535/) (2022)
- [MaxViT: Multi-Axis Vision Transformer](https://mlsanity.com/paper/2204.01697/) (2022)
- [Photorealistic Video Generation with Diffusion Models](https://mlsanity.com/paper/2312.06662/) (2023)
- [Diff-Instruct: A Universal Approach for Transferring Knowledge From Pre-trained Diffusion Models](https://mlsanity.com/paper/2305.18455/) (2023)
- [Score-Based Generative Modeling with Critically-Damped Langevin Diffusion](https://mlsanity.com/paper/2112.07068/) (2021)
- [Retinexformer: One-stage Retinex-based Transformer for Low-light Image Enhancement](https://mlsanity.com/paper/2303.06705/) (2023)
- [StyTr2: Image Style Transfer with Transformers](https://mlsanity.com/paper/2105.14576/) (2021)
- [Masked Discrimination for Self-Supervised Learning on Point Clouds](https://mlsanity.com/paper/2203.11183/) (2022)
- [Improving Sample Quality of Diffusion Models Using Self-Attention Guidance](https://mlsanity.com/paper/2210.00939/) (2022)
- [Anti-Oversmoothing in Deep Vision Transformers via the Fourier Domain Analysis: From Theory to Practice](https://mlsanity.com/paper/2203.05962/) (2022)
- [Multimodal Token Fusion for Vision Transformers](https://mlsanity.com/paper/2204.08721/) (2022)
- [A Survey of Visual Transformers](https://mlsanity.com/paper/2111.06091/) (2021)
- [Beyond Self-attention: External Attention using Two Linear Layers for Visual Tasks](https://mlsanity.com/paper/2105.02358/) (2021)
- [M3ViT: Mixture-of-Experts Vision Transformer for Efficient Multi-task Learning with Model-Accelerator Co-design](https://mlsanity.com/paper/2210.14793/) (2022)
- [Soft Truncation: A Universal Training Technique of Score-based Diffusion Model for High Precision Score Estimation](https://mlsanity.com/paper/2106.05527/) (2021)
- [DiffFit: Unlocking Transferability of Large Diffusion Models via Simple Parameter-Efficient Fine-Tuning](https://mlsanity.com/paper/2304.06648/) (2023)
- [Unified Visual Transformer Compression](https://mlsanity.com/paper/2203.08243/) (2022)
- [Structure-Aware Sparse-View X-ray 3D Reconstruction](https://mlsanity.com/paper/2311.10959/) (2023)
- [When Adversarial Training Meets Vision Transformers: Recipes from Training to Architecture](https://mlsanity.com/paper/2210.07540/) (2022)
- [MogaNet: Multi-order Gated Aggregation Network](https://mlsanity.com/paper/2211.03295/) (2022)
- [Your ViT is Secretly a Hybrid Discriminative-Generative Diffusion Model](https://mlsanity.com/paper/2208.07791/) (2022)
- [A Transformer-Based Feature Segmentation and Region Alignment Method For UAV-View Geo-Localization](https://mlsanity.com/paper/2201.09206/) (2022)
- [Countering Malicious DeepFakes: Survey, Battleground, and Horizon](https://mlsanity.com/paper/2103.00218/) (2021)
- [Multimodal Image Synthesis and Editing: The Generative AI Era](https://mlsanity.com/paper/2112.13592/) (2021)
- [Large-Vocabulary 3D Diffusion Model with Transformer](https://mlsanity.com/paper/2309.07920/) (2023)
- [Implicit Warping for Animation with Image Sets](https://mlsanity.com/paper/2210.01794/) (2022)

## Abstract (arXiv)

The recent explosive interest on transformers has suggested their potential to become powerful "universal" models for computer vision tasks, such as classification, detection, and segmentation. While those attempts mainly study the discriminative models, we explore transformers on some more notoriously difficult vision tasks, e.g., generative adversarial networks (GANs). Our goal is to conduct the first pilot study in building a GAN completely free of convolutions, using only pure transformer-based architectures. Our vanilla GAN architecture, dubbed TransGAN, consists of a memory-friendly transformer-based generator that progressively increases feature resolution, and correspondingly a multi-scale discriminator to capture simultaneously semantic contexts and low-level textures. On top of them, we introduce the new module of grid self-attention for alleviating the memory bottleneck further, in order to scale up TransGAN to high-resolution generation. We also develop a unique training recipe including a series of techniques that can mitigate the training instability issues of TransGAN, such as data augmentation, modified normalization, and relative position encoding. Our best architecture achieves highly competitive performance compared to current state-of-the-art GANs using convolutional backbones. Specifically, TransGAN sets new state-of-the-art inception score of 10.43 and FID of 18.28 on STL-10, outperforming StyleGAN-V2. When it comes to higher-resolution (e.g. 256 x 256) generation tasks, such as on CelebA-HQ and LSUN-Church, TransGAN continues to produce diverse visual examples with high fidelity and impressive texture details. In addition, we dive deep into the transformer-based generation models to understand how their behaviors differ from convolutional ones, by visualizing training dynamics. The code is available at https://github.com/VITA-Group/TransGAN.