Generative Adversarial Networks (GANs) rely heavily on large-scale training data for training high-quality image generation models. With limited training data, the GAN discriminator often suffers from severe overfitting which directly leads to degraded generation especially in generation diversity. Inspired by the recent advances in knowledge distillation (KD), we propose KD-DLGAN, a knowledge-distillation based generation framework that introduces pre-trained vision-language models for training effective data-limited generation models. KD-DLGAN consists of two innovative designs. The first is aggregated generative KD that mitigates the discriminator overfitting by challenging the discriminator with harder learning tasks and distilling more generalizable knowledge from the pre-trained models. The second is correlated generative KD that improves the generation diversity by distilling and preserving the diverse image-text correlation within the pre-trained models. Extensive experiments over multiple benchmarks show that KD-DLGAN achieves superior image generation with limited training data. In addition, KD-DLGAN complements the state-of-the-art with consistent and substantial performance gains.
@article{arxiv.2303.17158,
title = {KD-DLGAN: Data Limited Image Generation via Knowledge Distillation},
author = {Kaiwen Cui and Yingchen Yu and Fangneng Zhan and Shengcai Liao and Shijian Lu1 and Eric Xing},
journal= {arXiv preprint arXiv:2303.17158},
year = {2023}
}