Regularizing Discriminative Capability of CGANs for Semi-Supervised Generative Learning

Yi Liu, Guangchang Deng, Xiangping Zeng, Si Wu, Zhiwen Yu, Hau-San Wong; Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2020, pp. 5720-5729

Abstract


Semi-supervised generative learning aims to learn the underlying class-conditional distribution of partially labeled data. Generative Adversarial Networks (GANs) have led to promising progress in this task. However, it still needs to further explore the issue of imbalance between real labeled data and fake data in the adversarial learning process. To address this issue, we propose a regularization technique based on Random Regional Replacement (R^3-regularization) to facilitate the generative learning process. Specifically, we construct two types of between-class instances: cross-category ones and real-fake ones. These instances could be closer to the decision boundaries and are important for regularizing the classification and discriminative networks in our class-conditional GANs, which we refer to as R^3-CGAN. Better guidance from these two networks makes the generative network produce instances with class-specific information and high fidelity. We experiment with multiple standard benchmarks, and demonstrate that the R^3-regularization can lead to significant improvement in both classification and class-conditional image synthesis.

Related Material


[pdf] [video]
[bibtex]
@InProceedings{Liu_2020_CVPR,
author = {Liu, Yi and Deng, Guangchang and Zeng, Xiangping and Wu, Si and Yu, Zhiwen and Wong, Hau-San},
title = {Regularizing Discriminative Capability of CGANs for Semi-Supervised Generative Learning},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2020}
}