Zero-VAE-GAN: Generating Unseen Features for Generalized and Transductive Zero-Shot Learning
Zero-VAE-GAN: Generating Unseen Features for Generalized and Transductive Zero-Shot Learning
复制标题
Zero-VAE-GAN：为广义和传导性零样本学习生成看不见的特征
DOI：
10.1109/tip.2020.2964429
复制
发表时间：
2020-01-01
影响因子：
10.6
通讯作者：
Shao, Ling
中科院分区：
文献类型：
--
作者：
Gao, Rui;Hou, Xingsong;Shao, Ling
由于在训练过程中缺乏不可见的类数据，零次学习（Zero-shot learning，简称ZRL）是一项具有挑战性的任务。现有的作品试图通过一个共同的中间语义空间建立视觉空间和类空间之间的映射。现有方法的主要局限性是对所见类的强烈偏见，称为域转移问题，这导致在传统和广义CNOL任务中的性能不令人满意。为了应对这一挑战，我们建议通过为看不见的类生成特征来将CNOL转换为传统的监督学习。为此，提出了一种耦合变分自编码器（VAE）和生成对抗网络（GAN）的联合生成模型，称为Zero-VAE-GAN，以生成高质量的不可见特征。为了增强类别级别的区分能力，联合框架中引入了对抗性分类网络。此外，我们还提出了两种自训练策略来增加未标记的不可见特征，用于模型的转换扩展，在很大程度上解决了域转移问题。在五个标准基准测试和一个大规模数据集上的实验结果表明，我们的生成模型优于传统的最先进的方法，特别是广义的任务。此外，进一步改善的转导设置证明了所提出的自我训练策略的有效性。
Zero-shot learning (ZSL) is a challenging task due to the lack of unseen class data during training. Existing works attempt to establish a mapping between the visual and class spaces through a common intermediate semantic space. The main limitation of existing methods is the strong bias towards seen class, known as the domain shift problem, which leads to unsatisfactory performance in both conventional and generalized ZSL tasks. To tackle this challenge, we propose to convert ZSL to the conventional supervised learning by generating features for unseen classes. To this end, a joint generative model that couples variational autoencoder (VAE) and generative adversarial network (GAN), called Zero-VAE-GAN, is proposed to generate high-quality unseen features. To enhance the class-level discriminability, an adversarial categorization network is incorporated into the joint framework. Besides, we propose two self-training strategies to augment unlabeled unseen features for the transductive extension of our model, addressing the domain shift problem to a large extent. Experimental results on five standard benchmarks and a large-scale dataset demonstrate the superiority of our generative model over the state-of-the-art methods for conventional, especially generalized ZSL tasks. Moreover, the further improvement of the transductive setting demonstrates the effectiveness of the proposed self-training strategies.