Co-Sparse Textural Similarity for Image Segmentation

TL;DR

This paper introduces a novel co-sparse analysis-based algorithm for image segmentation that combines texture, color, and location information within a convex optimization framework, achieving competitive results.

Contribution

It presents a new textural similarity measure based on co-sparse representations and integrates it with color and location data using a Bayesian approach for improved segmentation.

Findings

Outperforms state-of-the-art interactive segmentation methods on Graz Benchmark.

Provides competitive results in unsupervised segmentation.

Efficiently parallelizable on graphics hardware.

Abstract

We propose an algorithm for segmenting natural images based on texture and color information, which leverages the co-sparse analysis model for image segmentation within a convex multilabel optimization framework. As a key ingredient of this method, we introduce a novel textural similarity measure, which builds upon the co-sparse representation of image patches. We propose a Bayesian approach to merge textural similarity with information about color and location. Combined with recently developed convex multilabel optimization methods this leads to an efficient algorithm for both supervised and unsupervised segmentation, which is easily parallelized on graphics hardware. The approach provides competitive results in unsupervised segmentation and outperforms state-of-the-art interactive segmentation methods on the Graz Benchmark.