Data InStance Prior (DISP) in Generative Adversarial Networks
This work addresses the critical problem of training GANs effectively with limited data, which is a common challenge for researchers and practitioners in various computer vision applications.
This paper tackles the problem of training Generative Adversarial Networks (GANs) with limited data, which typically leads to divergence and low-quality, undiverse samples. The authors propose a novel transfer learning method using a data instance prior derived from pre-trained networks, demonstrating superior image quality and diversity compared to existing state-of-the-art techniques across various GAN architectures and datasets.
Recent advances in generative adversarial networks (GANs) have shown remarkable progress in generating high-quality images. However, this gain in performance depends on the availability of a large amount of training data. In limited data regimes, training typically diverges, and therefore the generated samples are of low quality and lack diversity. Previous works have addressed training in low data setting by leveraging transfer learning and data augmentation techniques. We propose a novel transfer learning method for GANs in the limited data domain by leveraging informative data prior derived from self-supervised/supervised pre-trained networks trained on a diverse source domain. We perform experiments on several standard vision datasets using various GAN architectures (BigGAN, SNGAN, StyleGAN2) to demonstrate that the proposed method effectively transfers knowledge to domains with few target images, outperforming existing state-of-the-art techniques in terms of image quality and diversity. We also show the utility of data instance prior in large-scale unconditional image generation.