Dynamic Conceptional Contrastive Learning for Generalized Category Discovery
This addresses the open-world problem of clustering partially labeled data with both known and novel categories, particularly benefiting fine-grained visual recognition, though it appears incremental as it builds on self-supervised learning methods.
The paper tackles the generalized category discovery (GCD) problem by proposing Dynamic Conceptional Contrastive Learning (DCCL), which improves clustering accuracy by estimating underlying visual concepts and learning conceptional representation, achieving state-of-the-art results such as a 16.2% improvement on new classes for the CUB-200 dataset.
Generalized category discovery (GCD) is a recently proposed open-world problem, which aims to automatically cluster partially labeled data. The main challenge is that the unlabeled data contain instances that are not only from known categories of the labeled data but also from novel categories. This leads traditional novel category discovery (NCD) methods to be incapacitated for GCD, due to their assumption of unlabeled data are only from novel categories. One effective way for GCD is applying self-supervised learning to learn discriminate representation for unlabeled data. However, this manner largely ignores underlying relationships between instances of the same concepts (e.g., class, super-class, and sub-class), which results in inferior representation learning. In this paper, we propose a Dynamic Conceptional Contrastive Learning (DCCL) framework, which can effectively improve clustering accuracy by alternately estimating underlying visual conceptions and learning conceptional representation. In addition, we design a dynamic conception generation and update mechanism, which is able to ensure consistent conception learning and thus further facilitate the optimization of DCCL. Extensive experiments show that DCCL achieves new state-of-the-art performances on six generic and fine-grained visual recognition datasets, especially on fine-grained ones. For example, our method significantly surpasses the best competitor by 16.2% on the new classes for the CUB-200 dataset. Code is available at https://github.com/TPCD/DCCL.