Class-incremental Novel Class Discovery
This addresses the challenge of incremental learning for novel class discovery, which is incremental as it builds on rehearsal-based methods to prevent forgetting in a specific domain.
The paper tackles the problem of class-incremental Novel Class Discovery (class-iNCD), aiming to discover novel categories in unlabelled data using a pre-trained model on related labelled categories while preserving recognition of base classes, and demonstrates significant outperformance over state-of-the-art approaches on three benchmarks.
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