Transferable Contrastive Network for Generalized Zero-Shot Learning

Zero-shot learning (ZSL) is a challenging problem that aims to recognize the target categories without seen data, where semantic information is leveraged to transfer knowledge from some source classes. Although ZSL has made great progress in recent years, most existing approaches are easy to overfit the sources classes in generalized zero-shot learning (GZSL) task, which indicates that they learn little knowledge about target classes. To tackle such problem, we propose a novel Transferable Contrastive Network (TCN) that explicitly transfers knowledge from the source classes to the target classes. It automatically contrasts one image with different classes to judge whether they are consistent or not. By exploiting the class similarities to make knowledge transfer from source images to similar target classes, our approach is more robust to recognize the target images. Experiments on five benchmark datasets show the superiority of our approach for GZSL.

1. Introduction

Object recognition is one of the basic issues in computer vision. It has made great progress in recent years with the rapid development of deep learning approaches [18, 33, 30, 13], where large numbers of labeled images are required, such as ImageNet [28]. However, collecting and annotating large numbers of images are difficult, especially for fine-grained categories in specific domains. Moreover, such supervised learning approaches can only recognize a fixed number of categories, which is not flexible. In contrast, humans can learn from only a few samples or even recognize unseen objects. Therefore, learning visual classifiers with no need of human annotation is becoming a hot topic in recent years.