SDIP: Self-Reinforcement Deep Image Prior Framework for Image Processing

TL;DR

This paper introduces SDIP, an enhanced deep image prior framework that uses self-reinforcement to stabilize and improve image processing results by leveraging the correlation between network input and output during iterations.

Contribution

The paper proposes SDIP, a novel reinforcement learning-based extension of DIP that improves stability and performance in image processing tasks.

Findings

SDIP outperforms original DIP in multiple applications.

SDIP achieves more stable and accurate results.

Experimental results demonstrate significant improvements over state-of-the-art methods.

Abstract

Deep image prior (DIP) proposed in recent research has revealed the inherent trait of convolutional neural networks (CNN) for capturing substantial low-level image statistics priors. This framework efficiently addresses the inverse problems in image processing and has induced extensive applications in various domains. However, as the whole algorithm is initialized randomly, the DIP algorithm often lacks stability. Thus, this method still has space for further improvement. In this paper, we propose the self-reinforcement deep image prior (SDIP) as an improved version of the original DIP. We observed that the changes in the DIP networks' input and output are highly correlated during each iteration. SDIP efficiently utilizes this trait in a reinforcement learning manner, where the current iteration's output is utilized by a steering algorithm to update the network input for the next iteration, guiding the algorithm toward improved results. Experimental results across multiple applications demonstrate that our proposed SDIP framework offers improvement compared to the original DIP method and other state-of-the-art methods.