GenCo: Generative Co-training for Generative Adversarial Networks with Limited Data
This addresses the challenge of data-limited image generation for researchers and practitioners in computer vision, offering a novel co-training approach that can complement existing augmentation methods.
The paper tackles the problem of training Generative Adversarial Networks (GANs) with limited data, which often leads to discriminator over-fitting and sub-optimal models, by proposing GenCo, a Generative Co-training network that introduces multiple complementary discriminators to provide diverse supervision, resulting in superior generation performance as shown in extensive experiments over multiple benchmarks.
Training effective Generative Adversarial Networks (GANs) requires large amounts of training data, without which the trained models are usually sub-optimal with discriminator over-fitting. Several prior studies address this issue by expanding the distribution of the limited training data via massive and hand-crafted data augmentation. We handle data-limited image generation from a very different perspective. Specifically, we design GenCo, a Generative Co-training network that mitigates the discriminator over-fitting issue by introducing multiple complementary discriminators that provide diverse supervision from multiple distinctive views in training. We instantiate the idea of GenCo in two ways. The first way is Weight-Discrepancy Co-training (WeCo) which co-trains multiple distinctive discriminators by diversifying their parameters. The second way is Data-Discrepancy Co-training (DaCo) which achieves co-training by feeding discriminators with different views of the input images (e.g., different frequency components of the input images). Extensive experiments over multiple benchmarks show that GenCo achieves superior generation with limited training data. In addition, GenCo also complements the augmentation approach with consistent and clear performance gains when combined.