Despite recent progress in generative image modeling, successfully generating high-resolution, diverse samples from complex datasets such as ImageNet remains an elusive goal. To this end, we train Generative Adversarial Networks at the largest scale yet attempted, and study the instabilities specific to such scale. We find that applying orthogonal regularization to the generator renders it amenable to a simple "truncation trick," allowing fine control over the trade-off between sample fidelity and variety by reducing the variance of the Generator's input. Our modifications lead to models which set the new state of the art in class-conditional image synthesis. When trained on ImageNet at 128×128 resolution, our models (BigGANs) achieve an Inception Score (IS) of 166.5 and Fréchet Inception Distance (FID) of 7.4, improving over the previous best IS of 52.52 and FID of 18.65.
이 논문을 읽어서 무엇을 배울 수 있는지 알려주세요! 🤔
2014년 이후로 GAN에 관련한 다양한 논문들이 나왔는데, 이 논문들을 다 일일이 읽어보면 좋겠지만 그럴 수 없는 것이 현실입니다. 이 survey 논문을 통해서 GAN의 발전에 관한 history를 간략하게 이해를 기대해 볼 수 있지 않을까요?
어떤 내용의 논문인가요? 👋
2019년에 발표된 GAN에 관한 Survey 논문입니다.
Abstract (요약) 🕵🏻♂️
Despite recent progress in generative image modeling, successfully generating high-resolution, diverse samples from complex datasets such as ImageNet remains an elusive goal. To this end, we train Generative Adversarial Networks at the largest scale yet attempted, and study the instabilities specific to such scale. We find that applying orthogonal regularization to the generator renders it amenable to a simple "truncation trick," allowing fine control over the trade-off between sample fidelity and variety by reducing the variance of the Generator's input. Our modifications lead to models which set the new state of the art in class-conditional image synthesis. When trained on ImageNet at 128×128 resolution, our models (BigGANs) achieve an Inception Score (IS) of 166.5 and Fréchet Inception Distance (FID) of 7.4, improving over the previous best IS of 52.52 and FID of 18.65.
이 논문을 읽어서 무엇을 배울 수 있는지 알려주세요! 🤔
2014년 이후로 GAN에 관련한 다양한 논문들이 나왔는데, 이 논문들을 다 일일이 읽어보면 좋겠지만 그럴 수 없는 것이 현실입니다. 이 survey 논문을 통해서 GAN의 발전에 관한 history를 간략하게 이해를 기대해 볼 수 있지 않을까요?
같이 읽어보면 좋을 만한 글이나 이슈가 있을까요?
https://towardsdatascience.com/must-read-papers-on-gans-b665bbae3317
레퍼런스의 URL을 알려주세요! 🔗
https://www.semanticscholar.org/paper/Large-Scale-GAN-Training-for-High-Fidelity-Natural-Brock-Donahue/22aab110058ebbd198edb1f1e7b4f69fb13c0613#citing-papers