Zero-Shot Learning Based on Knowledge Sharing

TL;DR

This paper proposes a knowledge sharing approach combined with generative adversarial networks to improve zero-shot learning by enriching semantic features and generating more accurate pseudo visual features, leading to better classification performance.

Contribution

It introduces a novel knowledge sharing method to enhance semantic feature representation and employs GANs to generate realistic pseudo visual features for ZSL.

Findings

Consistent improvement on benchmark datasets

Enhanced semantic feature representation

Effective pseudo visual feature generation

Abstract

Zero-Shot Learning (ZSL) is an emerging research that aims to solve the classification problems with very few training data. The present works on ZSL mainly focus on the mapping of learning semantic space to visual space. It encounters many challenges that obstruct the progress of ZSL research. First, the representation of the semantic feature is inadequate to represent all features of the categories. Second, the domain drift problem still exists during the transfer from semantic space to visual space. In this paper, we introduce knowledge sharing (KS) to enrich the representation of semantic features. Based on KS, we apply a generative adversarial network to generate pseudo visual features from semantic features that are very close to the real visual features. Abundant experimental results from two benchmark datasets of ZSL show that the proposed approach has a consistent improvement.