Unsupervised Deep Learning Meets Chan-Vese Model

TL;DR

This paper introduces an unsupervised image segmentation method that combines the classical Chan-Vese model with deep neural networks, improving segmentation accuracy without needing pre-training.

Contribution

It integrates deep neural networks with the Chan-Vese model within a Bayesian framework, enabling unsupervised segmentation that handles violations of the piecewise constant assumption.

Findings

Outperforms existing unsupervised segmentation methods.

Effective in multi-phase and dataset-based segmentation tasks.

No pre-training required, only input images.

Abstract

The Chan-Vese (CV) model is a classic region-based method in image segmentation. However, its piecewise constant assumption does not always hold for practical applications. Many improvements have been proposed but the issue is still far from well solved. In this work, we propose an unsupervised image segmentation approach that integrates the CV model with deep neural networks, which significantly improves the original CV model's segmentation accuracy. Our basic idea is to apply a deep neural network that maps the image into a latent space to alleviate the violation of the piecewise constant assumption in image space. We formulate this idea under the classic Bayesian framework by approximating the likelihood with an evidence lower bound (ELBO) term while keeping the prior term in the CV model. Thus, our model only needs the input image itself and does not require pre-training from external datasets. Moreover, we extend the idea to multi-phase case and dataset based unsupervised image segmentation. Extensive experiments validate the effectiveness of our model and show that the proposed method is noticeably better than other unsupervised segmentation approaches.