Transferable Contrastive Network for Generalized Zero-Shot Learning

TL;DR

This paper introduces a Transferable Contrastive Network (TCN) that improves generalized zero-shot learning by explicitly transferring knowledge from source to target classes through contrastive learning, enhancing robustness and accuracy.

Contribution

The novel TCN model leverages class similarities and contrastive learning to better transfer knowledge in GZSL, addressing overfitting issues of previous methods.

Findings

Outperforms existing methods on five benchmark datasets

More robust recognition of target classes in GZSL

Effective knowledge transfer via class similarity exploitation

Abstract

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.