Hierarchical Compact Clustering Attention (COCA) for Unsupervised Object-Centric Learning

TL;DR

The paper introduces COCA, a hierarchical attention-based clustering method for unsupervised object-centric learning that effectively discovers and segments multiple objects in images without predefining object counts.

Contribution

It presents a novel clustering algorithm leveraging compactness within a hierarchical network, enabling high-quality unsupervised object segmentation without fixed object number constraints.

Findings

Achieves superior segmentation performance on six datasets.

Handles background segmentation better than competitors.

Operates effectively without predefining object count.

Abstract

We propose the Compact Clustering Attention (COCA) layer, an effective building block that introduces a hierarchical strategy for object-centric representation learning, while solving the unsupervised object discovery task on single images. COCA is an attention-based clustering module capable of extracting object-centric representations from multi-object scenes, when cascaded into a bottom-up hierarchical network architecture, referred to as COCA-Net. At its core, COCA utilizes a novel clustering algorithm that leverages the physical concept of compactness, to highlight distinct object centroids in a scene, providing a spatial inductive bias. Thanks to this strategy, COCA-Net generates high-quality segmentation masks on both the decoder side and, notably, the encoder side of its pipeline. Additionally, COCA-Net is not bound by a predetermined number of object masks that it generates and handles the segmentation of background elements better than its competitors. We demonstrate COCA-Net's segmentation performance on six widely adopted datasets, achieving superior or competitive results against the state-of-the-art models across nine different evaluation metrics.