Boosting Generative Zero-Shot Learning by Synthesizing Diverse Features with Attribute Augmentation

TL;DR

This paper introduces a novel approach to zero-shot learning that synthesizes diverse features by augmenting semantic attributes, leading to improved performance in generating more realistic visual features for unseen categories.

Contribution

The paper proposes a new framework that uses attribute augmentation to generate diverse visual features, addressing the limitation of complete attribute information in existing models.

Findings

Significant performance improvements on four benchmark datasets.

Enhanced diversity in generated features improves ZSL accuracy.

Outperforms state-of-the-art methods in zero-shot learning tasks.

Abstract

The recent advance in deep generative models outlines a promising perspective in the realm of Zero-Shot Learning (ZSL). Most generative ZSL methods use category semantic attributes plus a Gaussian noise to generate visual features. After generating unseen samples, this family of approaches effectively transforms the ZSL problem into a supervised classification scheme. However, the existing models use a single semantic attribute, which contains the complete attribute information of the category. The generated data also carry the complete attribute information, but in reality, visual samples usually have limited attributes. Therefore, the generated data from attribute could have incomplete semantics. Based on this fact, we propose a novel framework to boost ZSL by synthesizing diverse features. This method uses augmented semantic attributes to train the generative model, so as to simulate the real distribution of visual features. We evaluate the proposed model on four benchmark datasets, observing significant performance improvement against the state-of-the-art.