English

Cluster and Predict Latent Patches for Improved Masked Image Modeling

Computer Vision and Pattern Recognition 2025-07-01 v3 Artificial Intelligence

Abstract

Masked Image Modeling (MIM) offers a promising approach to self-supervised representation learning, however existing MIM models still lag behind the state-of-the-art. In this paper, we systematically analyze target representations, loss functions, and architectures, to introduce CAPI - a novel pure-MIM framework that relies on the prediction of latent clusterings. Our approach leverages a clustering-based loss, which is stable to train, and exhibits promising scaling properties. Our ViT-L backbone, CAPI, achieves 83.8% accuracy on ImageNet and 32.1% mIoU on ADE20K with simple linear probes, substantially outperforming previous MIM methods and approaching the performance of the current state-of-the-art, DINOv2. We release all our code and models.

Keywords

Cite

@article{arxiv.2502.08769,
  title  = {Cluster and Predict Latent Patches for Improved Masked Image Modeling},
  author = {Timothée Darcet and Federico Baldassarre and Maxime Oquab and Julien Mairal and Piotr Bojanowski},
  journal= {arXiv preprint arXiv:2502.08769},
  year   = {2025}
}

Comments

26 pages, 14 figures, accepted in TMLR 2025

R2 v1 2026-06-28T21:42:15.824Z