unMORE: Unsupervised Multi-Object Segmentation via Center-Boundary Reasoning

TL;DR

unMORE is a novel two-stage unsupervised multi-object segmentation method that explicitly learns object representations and reasoning, outperforming existing approaches on complex real-world images including crowded scenes.

Contribution

The paper introduces unMORE, a two-stage pipeline with explicit object-centric representation learning and a network-free reasoning module for unsupervised segmentation.

Findings

Outperforms all existing unsupervised methods on 6 real-world datasets.

Achieves state-of-the-art results on the COCO dataset.

Excels in crowded scenes where other methods fail.

Abstract

We study the challenging problem of unsupervised multi-object segmentation on single images. Existing methods, which rely on image reconstruction objectives to learn objectness or leverage pretrained image features to group similar pixels, often succeed only in segmenting simple synthetic objects or discovering a limited number of real-world objects. In this paper, we introduce unMORE, a novel two-stage pipeline designed to identify many complex objects in real-world images. The key to our approach involves explicitly learning three levels of carefully defined object-centric representations in the first stage. Subsequently, our multi-object reasoning module utilizes these learned object priors to discover multiple objects in the second stage. Notably, this reasoning module is entirely network-free and does not require human labels. Extensive experiments demonstrate that unMORE significantly outperforms all existing unsupervised methods across 6 real-world benchmark datasets, including the challenging COCO dataset, achieving state-of-the-art object segmentation results. Remarkably, our method excels in crowded images where all baselines collapse.