A regularization-based approach for unsupervised image segmentation

TL;DR

This paper introduces an unsupervised image segmentation method that over-segments images into superpixels, iteratively merges them based on a classifier and regularization, achieving competitive accuracy without supervision.

Contribution

It presents a novel regularization-based unsupervised segmentation algorithm that does not require prior knowledge of the number of segments and reduces oversegmentation.

Findings

Performs on par with state-of-the-art in precision

Reduces oversegmentation of objects

Operates without supervision

Abstract

We propose a novel unsupervised image segmentation algorithm, which aims to segment an image into several coherent parts. It requires no user input, no supervised learning phase and assumes an unknown number of segments. It achieves this by first over-segmenting the image into several hundred superpixels. These are iteratively joined on the basis of a discriminative classifier trained on color and texture information obtained from each superpixel. The output of the classifier is regularized by a Markov random field that lends more influence to neighbouring superpixels that are more similar. In each iteration, similar superpixels fall under the same label, until only a few coherent regions remain in the image. The algorithm was tested on a standard evaluation data set, where it performs on par with state-of-the-art algorithms in term of precision and greatly outperforms the state of the art by reducing the oversegmentation of the object of interest.