Conformal Bounds on Full-Reference Image Quality for Imaging Inverse Problems

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Wen, Jeffrey, Ahmad, Rizwan, Schniter, Philip
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866918020565172224
author Wen, Jeffrey
Ahmad, Rizwan
Schniter, Philip
author_facet Wen, Jeffrey
Ahmad, Rizwan
Schniter, Philip
contents In imaging inverse problems, we would like to know how close the recovered image is to the true image in terms of full-reference image quality (FRIQ) metrics like PSNR, SSIM, LPIPS, etc. This is especially important in safety-critical applications like medical imaging, where knowing that, say, the SSIM was poor could potentially avoid a costly misdiagnosis. But since we don't know the true image, computing FRIQ is non-trivial. In this work, we combine conformal prediction with approximate posterior sampling to construct bounds on FRIQ that are guaranteed to hold up to a user-specified error probability. We demonstrate our approach on image denoising and accelerated magnetic resonance imaging (MRI) problems. Code is available at https://github.com/jwen307/quality_uq.
format Preprint
id arxiv_https___arxiv_org_abs_2505_09528
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Conformal Bounds on Full-Reference Image Quality for Imaging Inverse Problems
Wen, Jeffrey
Ahmad, Rizwan
Schniter, Philip
Computer Vision and Pattern Recognition
In imaging inverse problems, we would like to know how close the recovered image is to the true image in terms of full-reference image quality (FRIQ) metrics like PSNR, SSIM, LPIPS, etc. This is especially important in safety-critical applications like medical imaging, where knowing that, say, the SSIM was poor could potentially avoid a costly misdiagnosis. But since we don't know the true image, computing FRIQ is non-trivial. In this work, we combine conformal prediction with approximate posterior sampling to construct bounds on FRIQ that are guaranteed to hold up to a user-specified error probability. We demonstrate our approach on image denoising and accelerated magnetic resonance imaging (MRI) problems. Code is available at https://github.com/jwen307/quality_uq.
title Conformal Bounds on Full-Reference Image Quality for Imaging Inverse Problems
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2505.09528