Self-supervised Object-Centric Learning for Videos

TL;DR

This paper introduces a fully unsupervised object-centric learning method for segmenting multiple objects in real-world videos, leveraging temporal relations and high-level features without additional modalities.

Contribution

It presents the first unsupervised approach that spatially binds objects to slots and relates them across frames for real-world video segmentation.

Findings

Successfully segments multiple complex objects in YouTube videos

Operates without supervised labels or additional modalities

Uses a novel masking and merging strategy for improved performance

Abstract

Unsupervised multi-object segmentation has shown impressive results on images by utilizing powerful semantics learned from self-supervised pretraining. An additional modality such as depth or motion is often used to facilitate the segmentation in video sequences. However, the performance improvements observed in synthetic sequences, which rely on the robustness of an additional cue, do not translate to more challenging real-world scenarios. In this paper, we propose the first fully unsupervised method for segmenting multiple objects in real-world sequences. Our object-centric learning framework spatially binds objects to slots on each frame and then relates these slots across frames. From these temporally-aware slots, the training objective is to reconstruct the middle frame in a high-level semantic feature space. We propose a masking strategy by dropping a significant portion of tokens in the feature space for efficiency and regularization. Additionally, we address over-clustering by merging slots based on similarity. Our method can successfully segment multiple instances of complex and high-variety classes in YouTube videos.