Exploring the Limits of Deep Image Clustering using Pretrained Models
Nikolas Adaloglou (HHU),* Felix Michels (HHU), Hamza Kalisch (HHU), Markus Kollmann (HHU)
The 34th British Machine Vision Conference
Abstract
We present a general methodology that learns to classify images without labels by leveraging pretrained feature extractors. Our approach involves self-distillation training of clustering heads, based on the fact that nearest neighbours in the pretrained feature space are likely to share the same label. We propose a novel objective that learns associations between image features by introducing a variant of pointwise mutual information together with instance weighting. We demonstrate that the proposed objective is able to attenuate the effect of false positive pairs while efficiently exploiting the structure in the pretrained feature space. As a result, we improve the clustering accuracy over $k$-means on $17$ different pretrained models by $6.1$\% and $12.2$\% on ImageNet and CIFAR100, respectively. Finally, using self-supervised vision transformers we achieve a clustering accuracy of $61.6$\% on ImageNet. The code is available at https://github.com/HHU-MMBS/TEMI-official-BMVC2023.Video
Citation
@inproceedings{Adaloglou_2023_BMVC,
author = {Nikolas Adaloglou and Felix Michels and Hamza Kalisch and Markus Kollmann},
title = {Exploring the Limits of Deep Image Clustering using Pretrained Models},
booktitle = {34th British Machine Vision Conference 2023, {BMVC} 2023, Aberdeen, UK, November 20-24, 2023},
publisher = {BMVA},
year = {2023},
url = {https://papers.bmvc2023.org/0297.pdf}
}
Copyright © 2023
The British Machine Vision Conference is organised by