Res-U2Net: Untrained Deep Learning for Phase Retrieval and Image Reconstruction

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Quero, Carlos Osorio, Leykam, Daniel, Ojeda, Irving Rondon
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909245219274752
author Quero, Carlos Osorio
Leykam, Daniel
Ojeda, Irving Rondon
author_facet Quero, Carlos Osorio
Leykam, Daniel
Ojeda, Irving Rondon
contents Conventional deep learning-based image reconstruction methods require a large amount of training data which can be hard to obtain in practice. Untrained deep learning methods overcome this limitation by training a network to invert a physical model of the image formation process. Here we present a novel untrained Res-U2Net model for phase retrieval. We use the extracted phase information to determine changes in an object's surface and generate a mesh representation of its 3D structure. We compare the performance of Res-U2Net phase retrieval against UNet and U2Net using images from the GDXRAY dataset.
format Preprint
id arxiv_https___arxiv_org_abs_2404_06657
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Res-U2Net: Untrained Deep Learning for Phase Retrieval and Image Reconstruction
Quero, Carlos Osorio
Leykam, Daniel
Ojeda, Irving Rondon
Image and Video Processing
Computer Vision and Pattern Recognition
Applied Physics
Optics
Conventional deep learning-based image reconstruction methods require a large amount of training data which can be hard to obtain in practice. Untrained deep learning methods overcome this limitation by training a network to invert a physical model of the image formation process. Here we present a novel untrained Res-U2Net model for phase retrieval. We use the extracted phase information to determine changes in an object's surface and generate a mesh representation of its 3D structure. We compare the performance of Res-U2Net phase retrieval against UNet and U2Net using images from the GDXRAY dataset.
title Res-U2Net: Untrained Deep Learning for Phase Retrieval and Image Reconstruction
topic Image and Video Processing
Computer Vision and Pattern Recognition
Applied Physics
Optics
url https://arxiv.org/abs/2404.06657