A Replica Inference Approach to Unsupervised Multi-Scale Image Segmentation

TL;DR

This paper introduces a novel unsupervised multi-scale image segmentation method based on replica inference in a Potts model, leveraging statistical mechanics concepts to identify stable image segments across resolutions.

Contribution

It applies replica inference and phase diagram analysis from statistical mechanics to improve multi-scale image segmentation, especially for camouflaged images.

Findings

Algorithm is fast and accurate, comparable to state-of-the-art methods.

Effective in detecting camouflaged and complex image structures.

Utilizes multiresolution correlations for robust segmentation.

Abstract

We apply a replica inference based Potts model method to unsupervised image segmentation on multiple scales. This approach was inspired by the statistical mechanics problem of "community detection" and its phase diagram. Specifically, the problem is cast as identifying tightly bound clusters ("communities" or "solutes") against a background or "solvent". Within our multiresolution approach, we compute information theory based correlations among multiple solutions ("replicas") of the same graph over a range of resolutions. Significant multiresolution structures are identified by replica correlations as manifest in information theory overlaps. With the aid of these correlations as well as thermodynamic measures, the phase diagram of the corresponding Potts model is analyzed both at zero and finite temperatures. Optimal parameters corresponding to a sensible unsupervised segmentation correspond to the "easy phase" of the Potts model. Our algorithm is fast and shown to be at least as accurate as the best algorithms to date and to be especially suited to the detection of camouflaged images.