Tada-DIP: Input-adaptive Deep Image Prior for One-shot 3D Image Reconstruction

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Bell, Evan, Liang, Shijun, Alkhouri, Ismail, Ravishankar, Saiprasad
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908691348848640
author Bell, Evan
Liang, Shijun
Alkhouri, Ismail
Ravishankar, Saiprasad
author_facet Bell, Evan
Liang, Shijun
Alkhouri, Ismail
Ravishankar, Saiprasad
contents Deep Image Prior (DIP) has recently emerged as a promising one-shot neural-network based image reconstruction method. However, DIP has seen limited application to 3D image reconstruction problems. In this work, we introduce Tada-DIP, a highly effective and fully 3D DIP method for solving 3D inverse problems. By combining input-adaptation and denoising regularization, Tada-DIP produces high-quality 3D reconstructions while avoiding the overfitting phenomenon that is common in DIP. Experiments on sparse-view X-ray computed tomography reconstruction validate the effectiveness of the proposed method, demonstrating that Tada-DIP produces much better reconstructions than training-data-free baselines and achieves reconstruction performance on par with a supervised network trained using a large dataset with fully-sampled volumes.
format Preprint
id arxiv_https___arxiv_org_abs_2512_03962
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Tada-DIP: Input-adaptive Deep Image Prior for One-shot 3D Image Reconstruction
Bell, Evan
Liang, Shijun
Alkhouri, Ismail
Ravishankar, Saiprasad
Image and Video Processing
Computer Vision and Pattern Recognition
Machine Learning
Deep Image Prior (DIP) has recently emerged as a promising one-shot neural-network based image reconstruction method. However, DIP has seen limited application to 3D image reconstruction problems. In this work, we introduce Tada-DIP, a highly effective and fully 3D DIP method for solving 3D inverse problems. By combining input-adaptation and denoising regularization, Tada-DIP produces high-quality 3D reconstructions while avoiding the overfitting phenomenon that is common in DIP. Experiments on sparse-view X-ray computed tomography reconstruction validate the effectiveness of the proposed method, demonstrating that Tada-DIP produces much better reconstructions than training-data-free baselines and achieves reconstruction performance on par with a supervised network trained using a large dataset with fully-sampled volumes.
title Tada-DIP: Input-adaptive Deep Image Prior for One-shot 3D Image Reconstruction
topic Image and Video Processing
Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2512.03962