Cross-Domain Evaluation of Few-Shot Classification Models: Natural Images vs. Histopathological Images

TL;DR

This paper evaluates the transferability of few-shot classification models between natural and histopathological images, revealing their strengths and limitations in cross-domain generalization and providing insights for optimizing performance.

Contribution

It provides a comprehensive cross-domain evaluation of few-shot models, highlighting their transferability and limitations between natural and medical image domains.

Findings

Models show varying transferability across domains

Performance depends on dataset and shot number

Insights into improving cross-domain few-shot learning

Abstract

In this study, we investigate the performance of few-shot classification models across different domains, specifically natural images and histopathological images. We first train several few-shot classification models on natural images and evaluate their performance on histopathological images. Subsequently, we train the same models on histopathological images and compare their performance. We incorporated four histopathology datasets and one natural images dataset and assessed performance across 5-way 1-shot, 5-way 5-shot, and 5-way 10-shot scenarios using a selection of state-of-the-art classification techniques. Our experimental results reveal insights into the transferability and generalization capabilities of few-shot classification models between diverse image domains. We analyze the strengths and limitations of these models in adapting to new domains and provide recommendations for optimizing their performance in cross-domain scenarios. This research contributes to advancing our understanding of few-shot learning in the context of image classification across diverse domains.