StarGAN v2: Diverse Image Synthesis for Multiple Domains
Yunjey Choi, Youngjung Uh, Jaejun Yoo, Jung-Woo Ha
Code Available — Be the first to reproduce this paper.
ReproduceCode
- github.com/clovaai/stargan-v2OfficialIn paperpytorch★ 3,610
- github.com/naver-ai/StyleMapGANpytorch★ 465
- github.com/mindslab-ai/hififacepytorch★ 383
- github.com/kunheek/style-aware-discriminatorpytorch★ 115
- github.com/karlchahine/neural-cover-selection-for-image-steganographypytorch★ 13
- github.com/SUPERSHOPxyz/stylegan3-gradientpytorch★ 1
- github.com/zzz2010/starganv2_paddlepytorch★ 0
- github.com/2023-MindSpore-4/Code7/tree/main/StarGANmindspore★ 0
- github.com/threeracha/Chuibbo-Flask-Serverpytorch★ 0
- github.com/sss20young/Chuibbo-Flask-Serverpytorch★ 0
Abstract
A good image-to-image translation model should learn a mapping between different visual domains while satisfying the following properties: 1) diversity of generated images and 2) scalability over multiple domains. Existing methods address either of the issues, having limited diversity or multiple models for all domains. We propose StarGAN v2, a single framework that tackles both and shows significantly improved results over the baselines. Experiments on CelebA-HQ and a new animal faces dataset (AFHQ) validate our superiority in terms of visual quality, diversity, and scalability. To better assess image-to-image translation models, we release AFHQ, high-quality animal faces with large inter- and intra-domain differences. The code, pretrained models, and dataset can be found at https://github.com/clovaai/stargan-v2.
Tasks
Benchmark Results
| Dataset | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| AFHQ | StarGAN v2 | LPIPS | 0.52 | — | Unverified |
| CelebA-HQ | StarGAN v2 | FID | 13.73 | — | Unverified |