JoJoGAN: One Shot Face Stylization (2021-12-22T00:00:00.000000Z)

TL;DR

Qualitative and quantitative evaluation show that JoJoGAN produces high quality high resolution images that vastly outperform the current state-of-the-art.

Abstract

A style mapper applies some fixed style to its input images (so, for example, taking faces to cartoons). This paper describes a simple procedure -- JoJoGAN -- to learn a style mapper from a single example of the style. JoJoGAN uses a GAN inversion procedure and StyleGAN's style-mixing property to produce a substantial paired dataset from a single example style. The paired dataset is then used to fine-tune a StyleGAN. An image can then be style mapped by GAN-inversion followed by the fine-tuned StyleGAN. JoJoGAN needs just one reference and as little as 30 seconds of training time. JoJoGAN can use extreme style references (say, animal faces) successfully. Furthermore, one can control what aspects of the style are used and how much of the style is applied. Qualitative and quantitative evaluation show that JoJoGAN produces high quality high resolution images that vastly outperform the current state-of-the-art.

Authors

Min Jin Chong

2 papers

David A. Forsyth

2 papers

TL;DR

Abstract

Authors

References41 items

StyleGAN of All Trades: Image Manipulation with Only Pretrained StyleGAN

BlendGAN: Implicitly GAN Blending for Arbitrary Stylized Face Generation

Mind the Gap: Domain Gap Control for Single Shot Domain Adaptation for Generative Adversarial Networks

StyleGAN-NADA

Retrieve in Style: Unsupervised Facial Feature Transfer and Retrieval

GANs N' Roses: Stable, Controllable, Diverse Image to Image Translation (works for videos too!)

GAN Prior Embedded Network for Blind Face Restoration in the Wild

Few-shot Image Generation via Cross-domain Correspondence

ReStyle: A Residual-Based StyleGAN Encoder via Iterative Refinement

Learning Transferable Visual Models From Natural Language Supervision

Designing an encoder for StyleGAN image manipulation

Towards Real-World Blind Face Restoration with Generative Facial Prior

Improved StyleGAN Embedding: Where are the Good Latents?

Few-shot Image Generation with Elastic Weight Consolidation

StyleSpace Analysis: Disentangled Controls for StyleGAN Image Generation

Resolution Dependent GAN Interpolation for Controllable Image Synthesis Between Domains

Few-Shot Adaptation of Generative Adversarial Networks

Closed-Form Factorization of Latent Semantics in GANs

InterFaceGAN: Interpreting the Disentangled Face Representation Learned by GANs

Deformable Style Transfer

PULSE: Self-Supervised Photo Upsampling via Latent Space Exploration of Generative Models

Freeze the Discriminator: a Simple Baseline for Fine-Tuning GANs

Analyzing and Improving the Image Quality of StyleGAN

Effectively Unbiased FID and Inception Score and Where to Find Them

Improving Style Transfer with Calibrated Metrics

A Style-Based Generator Architecture for Generative Adversarial Networks

Arbitrary Style Transfer With Style-Attentional Networks

Multimodal Unsupervised Image-to-Image Translation

ArcFace: Additive Angular Margin Loss for Deep Face Recognition

The Unreasonable Effectiveness of Deep Features as a Perceptual Metric

GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium

Universal Style Transfer via Feature Transforms

Arbitrary Style Transfer in Real-Time with Adaptive Instance Normalization

Image Style Transfer Using Convolutional Neural Networks

Improved Techniques for Training GANs

Perceptual Losses for Real-Time Style Transfer and Super-Resolution

Adam: A Method for Stochastic Optimization

Very Deep Convolutional Networks for Large-Scale Image Recognition

Image Analogies

Image quilting for texture synthesis and transfer

Unpaired image-to-image translation using cycle-consistent adversarial networks

Field of Study

Journal Information

Name

Volume

Venue Information

Name

Type

URL

Alternate Names