Cross Knowledge-based Generative Zero-Shot Learning Approach with Taxonomy Regularization

TL;DR

This paper introduces a generative zero-shot learning method that leverages cross knowledge learning and taxonomy regularization to improve recognition of unseen classes across different modalities and domains.

Contribution

It proposes a novel generative ZSL approach with CKL and TR, enhancing semantic-to-visual feature embedding and generalization to unseen classes.

Findings

Outperforms state-of-the-art methods on benchmark datasets

Achieves higher accuracy in ZSL image classification

Improves retrieval performance for unseen classes

Abstract

Although zero-shot learning (ZSL) has an inferential capability of recognizing new classes that have never been seen before, it always faces two fundamental challenges of the cross modality and crossdomain challenges. In order to alleviate these problems, we develop a generative network-based ZSL approach equipped with the proposed Cross Knowledge Learning (CKL) scheme and Taxonomy Regularization (TR). In our approach, the semantic features are taken as inputs, and the output is the synthesized visual features generated from the corresponding semantic features. CKL enables more relevant semantic features to be trained for semantic-to-visual feature embedding in ZSL, while Taxonomy Regularization (TR) significantly improves the intersections with unseen images with more generalized visual features generated from generative network. Extensive experiments on several benchmark datasets (i.e., AwA1, AwA2, CUB, NAB and aPY) show that our approach is superior to these state-of-the-art methods in terms of ZSL image classification and retrieval.