Class-incremental Novel Class Discovery

TL;DR

This paper introduces a novel class-incremental approach for discovering new categories in unlabeled data by leveraging pre-trained models, while preserving recognition of previously learned classes, outperforming existing methods.

Contribution

The paper proposes a new method combining feature prototypes, knowledge distillation, and self-training clustering for class-incremental novel class discovery.

Findings

Significantly outperforms state-of-the-art methods on benchmarks.

Effectively preserves base class recognition while discovering new classes.

Operates successfully in a class-incremental setting.

Abstract

We study the new task of class-incremental Novel Class Discovery (class-iNCD), which refers to the problem of discovering novel categories in an unlabelled data set by leveraging a pre-trained model that has been trained on a labelled data set containing disjoint yet related categories. Apart from discovering novel classes, we also aim at preserving the ability of the model to recognize previously seen base categories. Inspired by rehearsal-based incremental learning methods, in this paper we propose a novel approach for class-iNCD which prevents forgetting of past information about the base classes by jointly exploiting base class feature prototypes and feature-level knowledge distillation. We also propose a self-training clustering strategy that simultaneously clusters novel categories and trains a joint classifier for both the base and novel classes. This makes our method able to operate in a class-incremental setting. Our experiments, conducted on three common benchmarks, demonstrate that our method significantly outperforms state-of-the-art approaches. Code is available at https://github.com/OatmealLiu/class-iNCD