Multi-stream Fusion for Class Incremental Learning in Pill Image Classification

TL;DR

This paper introduces a multi-stream fusion framework with color guidance to improve class incremental learning in pill image classification, effectively addressing catastrophic forgetting in real-world healthcare applications.

Contribution

It proposes a novel incremental multi-stream fusion approach incorporating guidance streams, specifically color information, to enhance class incremental learning in pill image classification.

Findings

CG-IMIF outperforms state-of-the-art methods significantly.

The framework effectively mitigates catastrophic forgetting.

Experimental results on VAIPE-PCIL dataset demonstrate robustness.

Abstract

Classifying pill categories from real-world images is crucial for various smart healthcare applications. Although existing approaches in image classification might achieve a good performance on fixed pill categories, they fail to handle novel instances of pill categories that are frequently presented to the learning algorithm. To this end, a trivial solution is to train the model with novel classes. However, this may result in a phenomenon known as catastrophic forgetting, in which the system forgets what it learned in previous classes. In this paper, we address this challenge by introducing the class incremental learning (CIL) ability to traditional pill image classification systems. Specifically, we propose a novel incremental multi-stream intermediate fusion framework enabling incorporation of an additional guidance information stream that best matches the domain of the problem into various state-of-the-art CIL methods. From this framework, we consider color-specific information of pill images as a guidance stream and devise an approach, namely "Color Guidance with Multi-stream intermediate fusion"(CG-IMIF) for solving CIL pill image classification task. We conduct comprehensive experiments on real-world incremental pill image classification dataset, namely VAIPE-PCIL, and find that the CG-IMIF consistently outperforms several state-of-the-art methods by a large margin in different task settings. Our code, data, and trained model are available at https://github.com/vinuni-vishc/CG-IMIF.