A Feature Generator for Few-Shot Learning

TL;DR

This paper introduces a feature generator that synthesizes visual features from textual descriptions to improve few-shot learning accuracy, demonstrating significant performance gains over baseline methods.

Contribution

The paper presents a novel feature generator that uses textual descriptions to produce accurate visual features, enhancing few-shot learning performance.

Findings

10% accuracy improvement in 1-shot learning

5% accuracy improvement in 5-shot learning

Effective generation of class-specific features from textual data

Abstract

Few-shot learning (FSL) aims to enable models to recognize novel objects or classes with limited labelled data. Feature generators, which synthesize new data points to augment limited datasets, have emerged as a promising solution to this challenge. This paper investigates the effectiveness of feature generators in enhancing the embedding process for FSL tasks. To address the issue of inaccurate embeddings due to the scarcity of images per class, we introduce a feature generator that creates visual features from class-level textual descriptions. By training the generator with a combination of classifier loss, discriminator loss, and distance loss between the generated features and true class embeddings, we ensure the generation of accurate same-class features and enhance the overall feature representation. Our results show a significant improvement in accuracy over baseline methods, with our approach outperforming the baseline model by 10% in 1-shot and around 5% in 5-shot approaches. Additionally, both visual-only and visual + textual generators have also been tested in this paper. The code is publicly available at https://github.com/heethanjan/Feature-Generator-for-FSL.