# A Closer Look at Few-shot Image Generation

arXiv: https://arxiv.org/abs/2205.03805 · HTML: https://mlsanity.com/paper/2205.03805/

## Key facts

- Cited by 49 papers in the map (data through 14 Oct 2026).
- Builds on 45 of its 65 references that are in the map.
- First posted on arXiv 8 May 2022.

## Builds on (most cited first, 30 of 45)

- [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Representation Learning with Contrastive Predictive Coding](https://mlsanity.com/paper/1807.03748/) (2018)
- [Momentum Contrast for Unsupervised Visual Representation Learning](https://mlsanity.com/paper/1911.05722/) (2019)
- [Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization](https://mlsanity.com/paper/1610.02391/) (2016)
- [A Style-Based Generator Architecture for Generative Adversarial Networks](https://mlsanity.com/paper/1812.04948/) (2018)
- [Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks](https://mlsanity.com/paper/1703.03400/) (2017)
- [Overcoming catastrophic forgetting in neural networks](https://mlsanity.com/paper/1612.00796/) (2016)
- [Bootstrap your own latent: A new approach to self-supervised Learning](https://mlsanity.com/paper/2006.07733/) (2020)
- [Image-to-Image Translation with Conditional Adversarial Networks](https://mlsanity.com/paper/1611.07004/) (2016)
- [Analyzing and Improving the Image Quality of StyleGAN](https://mlsanity.com/paper/1912.04958/) (2019)
- [Prototypical Networks for Few-shot Learning](https://mlsanity.com/paper/1703.05175/) (2017)
- [Exploring Simple Siamese Representation Learning](https://mlsanity.com/paper/2011.10566/) (2020)
- [Unsupervised Learning of Visual Features by Contrasting Cluster Assignments](https://mlsanity.com/paper/2006.09882/) (2020)
- [Large Scale GAN Training for High Fidelity Natural Image Synthesis](https://mlsanity.com/paper/1809.11096/) (2018)
- [Unsupervised Feature Learning via Non-Parametric Instance-level Discrimination](https://mlsanity.com/paper/1805.01978/) (2018)
- [LSUN: Construction of a Large-scale Image Dataset using Deep Learning with Humans in the Loop](https://mlsanity.com/paper/1506.03365/) (2015)
- [Training Generative Adversarial Networks with Limited Data](https://mlsanity.com/paper/2006.06676/) (2020)
- [Learning to Compare: Relation Network for Few-Shot Learning](https://mlsanity.com/paper/1711.06025/) (2017)
- [CNN-generated images are surprisingly easy to spot... for now](https://mlsanity.com/paper/1912.11035/) (2019)
- [Self-Supervised Learning of Pretext-Invariant Representations](https://mlsanity.com/paper/1912.01991/) (2019)
- [Contrastive Learning for Unpaired Image-to-Image Translation](https://mlsanity.com/paper/2007.15651/) (2020)
- [A Survey on Contrastive Self-supervised Learning](https://mlsanity.com/paper/2011.00362/) (2020)
- [Which Tasks Should Be Learned Together in Multi-task Learning?](https://mlsanity.com/paper/1905.07553/) (2019)
- [Differentiable Augmentation for Data-Efficient GAN Training](https://mlsanity.com/paper/2006.10738/) (2020)
- [Data Augmentation Generative Adversarial Networks](https://mlsanity.com/paper/1711.04340/) (2017)
- [Few-shot Image Generation via Cross-domain Correspondence](https://mlsanity.com/paper/2104.06820/) (2021)
- [Consistency Regularization for Generative Adversarial Networks](https://mlsanity.com/paper/1910.12027/) (2019)
- [Transferring GANs: generating images from limited data](https://mlsanity.com/paper/1805.01677/) (2018)
- [Freeze the Discriminator: a Simple Baseline for Fine-Tuning GANs](https://mlsanity.com/paper/2002.10964/) (2020)

## Built on by (most cited first, 30 of 49)

- [A Recipe for Watermarking Diffusion Models](https://mlsanity.com/paper/2303.10137/) (2023)
- [Task-specific Fine-tuning via Variational Information Bottleneck for Weakly-supervised Pathology Whole Slide Image Classification](https://mlsanity.com/paper/2303.08446/) (2023)
- [Few-shot Image Generation via Adaptation-Aware Kernel Modulation](https://mlsanity.com/paper/2210.16559/) (2022)
- [Discovering Transferable Forensic Features for CNN-generated Images Detection](https://mlsanity.com/paper/2208.11342/) (2022)
- [Towards Diverse and Faithful One-shot Adaption of Generative Adversarial Networks](https://mlsanity.com/paper/2207.08736/) (2022)
- [Few-shot Image Generation with Diffusion Models](https://mlsanity.com/paper/2211.03264/) (2022)
- [Exploring Incompatible Knowledge Transfer in Few-shot Image Generation](https://mlsanity.com/paper/2304.07574/) (2023)
- [A Survey on Generative Modeling with Limited Data, Few Shots, and Zero Shot](https://mlsanity.com/paper/2307.14397/) (2023)
- [Domain Re-Modulation for Few-Shot Generative Domain Adaptation](https://mlsanity.com/paper/2302.02550/) (2023)
- [Unlearnable Examples Give a False Sense of Security: Piercing through Unexploitable Data with Learnable Examples](https://mlsanity.com/paper/2305.09241/) (2023)
- [DomainStudio: Fine-Tuning Diffusion Models for Domain-Driven Image Generation using Limited Data](https://mlsanity.com/paper/2306.14153/) (2023)
- [Scenimefy: Learning to Craft Anime Scene via Semi-Supervised Image-to-Image Translation](https://mlsanity.com/paper/2308.12968/) (2023)
- [Deformable One-shot Face Stylization via DINO Semantic Guidance](https://mlsanity.com/paper/2403.00459/) (2024)
- [Few-shot Image Generation via Masked Discrimination](https://mlsanity.com/paper/2210.15194/) (2022)
- [WeditGAN: Few-Shot Image Generation via Latent Space Relocation](https://mlsanity.com/paper/2305.06671/) (2023)
- [LFS-GAN: Lifelong Few-Shot Image Generation](https://mlsanity.com/paper/2308.11917/) (2023)
- [Transfer Learning for Diffusion Models](https://mlsanity.com/paper/2405.16876/) (2024)
- [FS-BAN: Born-Again Networks for Domain Generalization Few-Shot Classification](https://mlsanity.com/paper/2208.10930/) (2022)
- [Image Synthesis under Limited Data: A Survey and Taxonomy](https://mlsanity.com/paper/2307.16879/) (2023)
- [Generative Artificial Intelligence Meets Synthetic Aperture Radar: A Survey](https://mlsanity.com/paper/2411.05027/) (2024)
- [Stable Attribute Group Editing for Reliable Few-shot Image Generation](https://mlsanity.com/paper/2302.00179/) (2023)
- [One-for-All: Towards Universal Domain Translation with a Single StyleGAN](https://mlsanity.com/paper/2310.14222/) (2023)
- [Peer is Your Pillar: A Data-unbalanced Conditional GANs for Few-shot Image Generation](https://mlsanity.com/paper/2311.08217/) (2023)
- [Few-shot Semantic Image Synthesis with Class Affinity Transfer](https://mlsanity.com/paper/2304.02321/) (2023)
- [Improving the Effectiveness of Deep Generative Data](https://mlsanity.com/paper/2311.03959/) (2023)
- [CFTS-GAN: Continual Few-Shot Teacher Student for Generative Adversarial Networks](https://mlsanity.com/paper/2410.14749/) (2024)
- [GCA-3D: Towards Generalized and Consistent Domain Adaptation of 3D Generators](https://mlsanity.com/paper/2412.15491/) (2024)
- [Smoothness Similarity Regularization for Few-Shot GAN Adaptation](https://mlsanity.com/paper/2308.09717/) (2023)
- [Efficient Transfer Learning in Diffusion Models via Adversarial Noise](https://mlsanity.com/paper/2308.11948/) (2023)
- [UGC: Unified GAN Compression for Efficient Image-to-Image Translation](https://mlsanity.com/paper/2309.09310/) (2023)

## Abstract (arXiv)

Modern GANs excel at generating high quality and diverse images. However, when transferring the pretrained GANs on small target data (e.g., 10-shot), the generator tends to replicate the training samples. Several methods have been proposed to address this few-shot image generation task, but there is a lack of effort to analyze them under a unified framework. As our first contribution, we propose a framework to analyze existing methods during the adaptation. Our analysis discovers that while some methods have disproportionate focus on diversity preserving which impede quality improvement, all methods achieve similar quality after convergence. Therefore, the better methods are those that can slow down diversity degradation. Furthermore, our analysis reveals that there is still plenty of room to further slow down diversity degradation. Informed by our analysis and to slow down the diversity degradation of the target generator during adaptation, our second contribution proposes to apply mutual information (MI) maximization to retain the source domain's rich multi-level diversity information in the target domain generator. We propose to perform MI maximization by contrastive loss (CL), leverage the generator and discriminator as two feature encoders to extract different multi-level features for computing CL. We refer to our method as Dual Contrastive Learning (DCL). Extensive experiments on several public datasets show that, while leading to a slower diversity-degrading generator during adaptation, our proposed DCL brings visually pleasant quality and state-of-the-art quantitative performance. Project Page: yunqing-me.github.io/A-Closer-Look-at-FSIG.