Unsupervised video-based object-centric learning is a promising avenue to learn structured representations from large, unlabeled video collections, but previous approaches have only managed to scale to real-world datasets in restricted domains. Recently, it was shown that the reconstruction of pre-trained self-supervised features leads to object-centric representations on unconstrained real-world image datasets. Building on this approach, we propose a novel way to use such pre-trained features in the form of a temporal feature similarity loss. This loss encodes semantic and temporal correlations between image patches and is a natural way to introduce a motion bias for object discovery. We demonstrate that this loss leads to state-of-the-art performance on the challenging synthetic MOVi datasets. When used in combination with the feature reconstruction loss, our model is the first object-centric video model that scales to unconstrained video datasets such as YouTube-VIS.
If you find this work useful, please cite our paper
@inproceedings{zadaianchuk2023objectcentric,
title={Object-Centric Learning for Real-World Videos by Predicting Temporal Feature Similarities},
author={Zadaianchuk, Andrii and Seitzer, Maximilian and Martius, Georg},
booktitle={Thirty-seventh Conference on Neural Information Processing Systems (NeurIPS 2023)},
year={2023},
}