Extract More from Less: Efficient Fine-Grained Visual Recognition in Low-Data Regimes

TL;DR

This paper introduces AD-Net, a novel framework that improves fine-grained image classification in low-data regimes by combining augmentation and self-distillation, achieving significant accuracy gains without extra inference cost.

Contribution

The authors propose AD-Net, a low-data fine-grained classification framework that leverages augmentation and self-distillation, demonstrating substantial accuracy improvements and architecture independence.

Findings

Up to 45% accuracy increase over ResNet-50 with minimal data.

Consistent improvements over traditional fine-tuning and state-of-the-art low-data methods.

No additional inference cost introduced by the framework.

Abstract

The emerging task of fine-grained image classification in low-data regimes assumes the presence of low inter-class variance and large intra-class variation along with a highly limited amount of training samples per class. However, traditional ways of separately dealing with fine-grained categorisation and extremely scarce data may be inefficient under both these harsh conditions presented together. In this paper, we present a novel framework, called AD-Net, aiming to enhance deep neural network performance on this challenge by leveraging the power of Augmentation and Distillation techniques. Specifically, our approach is designed to refine learned features through self-distillation on augmented samples, mitigating harmful overfitting. We conduct comprehensive experiments on popular fine-grained image classification benchmarks where our AD-Net demonstrates consistent improvement over traditional fine-tuning and state-of-the-art low-data techniques. Remarkably, with the smallest data available, our framework shows an outstanding relative accuracy increase of up to 45 % compared to standard ResNet-50 and up to 27 % compared to the closest SOTA runner-up. We emphasise that our approach is practically architecture-independent and adds zero extra cost at inference time. Additionally, we provide an extensive study on the impact of every framework's component, highlighting the importance of each in achieving optimal performance. Source code and trained models are publicly available at github.com/demidovd98/fgic_lowd.