Cluster and Predict Latent Patches for Improved Masked Image Modeling
This work addresses the performance gap in MIM for computer vision researchers, offering a novel method that is incremental but shows strong gains.
The paper tackles the problem of improving Masked Image Modeling (MIM) for self-supervised representation learning by introducing CAPI, a framework that predicts latent clusterings, achieving 83.8% accuracy on ImageNet and 32.1% mIoU on ADE20K with linear probes, which outperforms previous MIM methods and approaches DINOv2's performance.
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.