Saved in:
Bibliographic Details
Main Authors: Kawachi, Hodaka, Neto, Jose Reinaldo Cunha Santos A. V. Silva, Yagi, Yasushi, Nagahara, Hajime, Nakamura, Tomoya
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2509.17427
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916971163942912
author Kawachi, Hodaka
Neto, Jose Reinaldo Cunha Santos A. V. Silva
Yagi, Yasushi
Nagahara, Hajime
Nakamura, Tomoya
author_facet Kawachi, Hodaka
Neto, Jose Reinaldo Cunha Santos A. V. Silva
Yagi, Yasushi
Nagahara, Hajime
Nakamura, Tomoya
contents We propose a single-snapshot depth-from-defocus (DFD) reconstruction method for coded-aperture imaging that replaces hand-crafted priors with a learned diffusion prior used purely as regularization. Our optimization framework enforces measurement consistency via a differentiable forward model while guiding solutions with the diffusion prior in the denoised image domain, yielding higher accuracy and stability than classical optimization. Unlike U-Net-style regressors, our approach requires no paired defocus-RGBD training data and does not tie training to a specific camera configuration. Experiments on comprehensive simulations and a prototype camera demonstrate consistently strong RGBD reconstructions across noise levels, outperforming both U-Net baselines and a classical coded-aperture DFD method.
format Preprint
id arxiv_https___arxiv_org_abs_2509_17427
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Single-Image Depth from Defocus with Coded Aperture and Diffusion Posterior Sampling
Kawachi, Hodaka
Neto, Jose Reinaldo Cunha Santos A. V. Silva
Yagi, Yasushi
Nagahara, Hajime
Nakamura, Tomoya
Computer Vision and Pattern Recognition
We propose a single-snapshot depth-from-defocus (DFD) reconstruction method for coded-aperture imaging that replaces hand-crafted priors with a learned diffusion prior used purely as regularization. Our optimization framework enforces measurement consistency via a differentiable forward model while guiding solutions with the diffusion prior in the denoised image domain, yielding higher accuracy and stability than classical optimization. Unlike U-Net-style regressors, our approach requires no paired defocus-RGBD training data and does not tie training to a specific camera configuration. Experiments on comprehensive simulations and a prototype camera demonstrate consistently strong RGBD reconstructions across noise levels, outperforming both U-Net baselines and a classical coded-aperture DFD method.
title Single-Image Depth from Defocus with Coded Aperture and Diffusion Posterior Sampling
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2509.17427