Simple Unsupervised Object-Centric Learning for Complex and Naturalistic Videos

TL;DR

STEVE is a simple yet effective unsupervised object-centric learning model for complex naturalistic videos, achieving significant improvements without added complexity or supervision.

Contribution

It introduces a straightforward transformer-based architecture for object-centric learning in videos, capable of handling complex scenes without additional supervision.

Findings

Outperforms previous methods on complex naturalistic videos

Uses a simple architecture without extra supervision

Achieves significant improvements in object-centric learning

Abstract

Unsupervised object-centric learning aims to represent the modular, compositional, and causal structure of a scene as a set of object representations and thereby promises to resolve many critical limitations of traditional single-vector representations such as poor systematic generalization. Although there have been many remarkable advances in recent years, one of the most critical problems in this direction has been that previous methods work only with simple and synthetic scenes but not with complex and naturalistic images or videos. In this paper, we propose STEVE, an unsupervised model for object-centric learning in videos. Our proposed model makes a significant advancement by demonstrating its effectiveness on various complex and naturalistic videos unprecedented in this line of research. Interestingly, this is achieved by neither adding complexity to the model architecture nor introducing a new objective or weak supervision. Rather, it is achieved by a surprisingly simple architecture that uses a transformer-based image decoder conditioned on slots and the learning objective is simply to reconstruct the observation. Our experiment results on various complex and naturalistic videos show significant improvements compared to the previous state-of-the-art.