Prior Knowledge about Attributes: Learning a More Effective Potential Space for Zero-Shot Recognition

TL;DR

This paper introduces ACPSG, a graph convolution network-based model that leverages attribute correlations to generate a discriminative potential space, significantly improving zero-shot recognition accuracy.

Contribution

The paper proposes a novel attribute correlation potential space generation method that enhances zero-shot learning by modeling attribute correlations with graph convolution networks.

Findings

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

Effective in both conventional and generalized ZSL scenarios

Improves class discrimination by modeling attribute correlations

Abstract

Zero-shot learning (ZSL) aims to recognize unseen classes accurately by learning seen classes and known attributes, but correlations in attributes were ignored by previous study which lead to classification results confused. To solve this problem, we build an Attribute Correlation Potential Space Generation (ACPSG) model which uses a graph convolution network and attribute correlation to generate a more discriminating potential space. Combining potential discrimination space and user-defined attribute space, we can better classify unseen classes. Our approach outperforms some existing state-of-the-art methods on several benchmark datasets, whether it is conventional ZSL or generalized ZSL.