LOCATE: Self-supervised Object Discovery via Flow-guided Graph-cut and Bootstrapped Self-training

TL;DR

LOCATE is a self-supervised method for object discovery that combines motion and appearance cues via graph cuts and self-training, achieving state-of-the-art results on various benchmarks without human labels.

Contribution

It introduces a novel flow-guided graph cut and bootstrapped self-training framework for unsupervised object segmentation, surpassing previous methods.

Findings

Achieves state-of-the-art performance on multiple benchmarks.

Effectively transfers to in-the-wild images.

Component ablation validates design choices.

Abstract

Learning object segmentation in image and video datasets without human supervision is a challenging problem. Humans easily identify moving salient objects in videos using the gestalt principle of common fate, which suggests that what moves together belongs together. Building upon this idea, we propose a self-supervised object discovery approach that leverages motion and appearance information to produce high-quality object segmentation masks. Specifically, we redesign the traditional graph cut on images to include motion information in a linear combination with appearance information to produce edge weights. Remarkably, this step produces object segmentation masks comparable to the current state-of-the-art on multiple benchmarks. To further improve performance, we bootstrap a segmentation network trained on these preliminary masks as pseudo-ground truths to learn from its own outputs via self-training. We demonstrate the effectiveness of our approach, named LOCATE, on multiple standard video object segmentation, image saliency detection, and object segmentation benchmarks, achieving results on par with and, in many cases surpassing state-of-the-art methods. We also demonstrate the transferability of our approach to novel domains through a qualitative study on in-the-wild images. Additionally, we present extensive ablation analysis to support our design choices and highlight the contribution of each component of our proposed method.