SINCO: A Novel structural regularizer for image compression using implicit neural representations

TL;DR

SINCO introduces a structural regularizer for implicit neural representations, enhancing image compression quality by enforcing structural consistency through segmentation masks, validated on brain MRI images.

Contribution

The paper proposes a novel structural regularizer for INR-based image compression that improves image quality by leveraging segmentation consistency.

Findings

SINCO outperforms recent INR methods on brain MRI images.

Structural regularization improves image quality in INR compression.

Segmentation-based regularization enhances structural fidelity of compressed images.

Abstract

Implicit neural representations (INR) have been recently proposed as deep learning (DL) based solutions for image compression. An image can be compressed by training an INR model with fewer weights than the number of image pixels to map the coordinates of the image to corresponding pixel values. While traditional training approaches for INRs are based on enforcing pixel-wise image consistency, we propose to further improve image quality by using a new structural regularizer. We present structural regularization for INR compression (SINCO) as a novel INR method for image compression. SINCO imposes structural consistency of the compressed images to the groundtruth by using a segmentation network to penalize the discrepancy of segmentation masks predicted from compressed images. We validate SINCO on brain MRI images by showing that it can achieve better performance than some recent INR methods.