Preserving Semantic Relations for Zero-Shot Learning

TL;DR

This paper enhances zero-shot learning by preserving semantic relations in attribute space, improving recognition accuracy and enabling semantic inference without attribute data.

Contribution

It introduces a novel method to incorporate attribute relations into the embedding space, boosting zero-shot learning performance.

Findings

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

Effective in both standard and generalized zero-shot settings

Enables semantic inference without attribute data

Abstract

Zero-shot learning has gained popularity due to its potential to scale recognition models without requiring additional training data. This is usually achieved by associating categories with their semantic information like attributes. However, we believe that the potential offered by this paradigm is not yet fully exploited. In this work, we propose to utilize the structure of the space spanned by the attributes using a set of relations. We devise objective functions to preserve these relations in the embedding space, thereby inducing semanticity to the embedding space. Through extensive experimental evaluation on five benchmark datasets, we demonstrate that inducing semanticity to the embedding space is beneficial for zero-shot learning. The proposed approach outperforms the state-of-the-art on the standard zero-shot setting as well as the more realistic generalized zero-shot setting. We also demonstrate how the proposed approach can be useful for making approximate semantic inferences about an image belonging to a category for which attribute information is not available.