Using Deep Image Prior to Assist Variational Selective Segmentation Deep Learning Algorithms

TL;DR

This paper combines Deep Image Prior with variational segmentation algorithms to leverage implicit regularisation while enabling future image prediction, enhancing traditional methods with deep learning insights.

Contribution

It introduces a novel approach that integrates Deep Image Prior into variational segmentation, allowing for implicit regularisation and future image prediction.

Findings

Improved segmentation quality with implicit regularisation.

Ability to predict future images using the combined method.

Competitive performance compared to existing approaches.

Abstract

Variational segmentation algorithms require a prior imposed in the form of a regularisation term to enforce smoothness of the solution. Recently, it was shown in the Deep Image Prior work that the explicit regularisation in a model can be removed and replaced by the implicit regularisation captured by the architecture of a neural network. The Deep Image Prior approach is competitive, but is only tailored to one specific image and does not allow us to predict future images. We propose to incorporate the ideas from Deep Image Prior into a more traditional learning algorithm to allow us to use the implicit regularisation offered by the Deep Image Prior, but still be able to predict future images.