A Modular and Robust Physics-Based Approach for Lensless Image Reconstruction

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Perron, Yohann, Bezzam, Eric, Vetterli, Martin
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866916410440024064
author Perron, Yohann
Bezzam, Eric
Vetterli, Martin
author_facet Perron, Yohann
Bezzam, Eric
Vetterli, Martin
contents In this paper, we present a modular approach for reconstructing lensless measurements. It consists of three components: a newly-proposed pre-processor, a physics-based camera inverter to undo the multiplexing of lensless imaging, and a post-processor. The pre- and post-processors address noise and artifacts unique to lensless imaging before and after camera inversion respectively. By training the three components end-to-end, we obtain a 1.9 dB increase in PSNR and a 14% relative improvement in a perceptual image metric (LPIPS) with respect to previously proposed physics-based methods. We also demonstrate how the proposed pre-processor provides more robustness to input noise, and how an auxiliary loss can improve interpretability.
format Preprint
id arxiv_https___arxiv_org_abs_2403_00537
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle A Modular and Robust Physics-Based Approach for Lensless Image Reconstruction
Perron, Yohann
Bezzam, Eric
Vetterli, Martin
Image and Video Processing
In this paper, we present a modular approach for reconstructing lensless measurements. It consists of three components: a newly-proposed pre-processor, a physics-based camera inverter to undo the multiplexing of lensless imaging, and a post-processor. The pre- and post-processors address noise and artifacts unique to lensless imaging before and after camera inversion respectively. By training the three components end-to-end, we obtain a 1.9 dB increase in PSNR and a 14% relative improvement in a perceptual image metric (LPIPS) with respect to previously proposed physics-based methods. We also demonstrate how the proposed pre-processor provides more robustness to input noise, and how an auxiliary loss can improve interpretability.
title A Modular and Robust Physics-Based Approach for Lensless Image Reconstruction
topic Image and Video Processing
url https://arxiv.org/abs/2403.00537