Pohang University of Science and Technology (POSTECH)

We address the problem of generalized category discovery (GCD) that aims to partition a partially labeled collection of images; only a small part of the collection is labeled and the total number of target classes is unknown. To address this generalized image clustering problem, we revisit the mean-shift algorithm, i.e., a classic, powerful technique for mode seeking, and incorporate it into a contrastive learning framework. The proposed method, dubbed Contrastive Mean-Shift (CMS) learning, trains an image encoder to produce representations with better clustering properties by an iterative process of mean shift and contrastive update. Experiments demonstrate that our method, both in settings with and without the total number of clusters being known, achieves state-of-the-art performance on six public GCD benchmarks without bells and whistles.

Comparison with the state of the arts on GCD using

kNN retrieved images of the initial embedding $\boldsymbol{v}$ and mean-shifted embedding $\boldsymbol{z}$ on CUB-200-2011. Green denotes the correct class and red an incorrect class.

```
@inproceedings{choi2024contrastive,
title={Contrastive Mean-Shift Learning for Generalized Category Discovery},
author={Choi, Sua and Kang, Dahyun and Cho, Minsu},
booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
year={2024}
}
```