dorsal/arxiv
View SchemaA Single-Parameter Factor-Graph Image Prior
| Authors | Tianyang Wang, Ender Konukoglu, Hans-Andrea Loeliger |
|---|---|
| Categories | |
| ArXiv ID | 2601.08749vv2 |
| URL | https://arxiv.org/abs/2601.08749 |
| License | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ |
Abstract
We propose a novel piecewise smooth image model with piecewise constant local parameters that are automatically adapted to each image. Technically, the model is formulated in terms of factor graphs with NUP (normal with unknown parameters) priors, and the pertinent computations amount to iterations of conjugate-gradient steps and Gaussian message passing. The proposed model and algorithms are demonstrated with applications to denoising and contrast enhancement.
{
"annotation_id": "b407c507-6bac-4a29-8942-6d60f3ea65dc",
"date_created": "2026-02-17T05:53:15.900000Z",
"date_modified": "2026-02-17T05:53:15.900000Z",
"file_hash": "4dcbd09b6174a76401c380cd8a232ca218a8a4e206eac246d9fee29a901445a0",
"private": false,
"record": {
"abstract": "We propose a novel piecewise smooth image model with piecewise constant local parameters that are automatically adapted to each image. Technically, the model is formulated in terms of factor graphs with NUP (normal with unknown parameters) priors, and the pertinent computations amount to iterations of conjugate-gradient steps and Gaussian message passing. The proposed model and algorithms are demonstrated with applications to denoising and contrast enhancement.",
"arxiv_id": "2601.08749",
"authors": [
"Tianyang Wang",
"Ender Konukoglu",
"Hans-Andrea Loeliger"
],
"categories": [
"eess.IV",
"cs.CV",
"eess.SP"
],
"license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
"title": "A Single-Parameter Factor-Graph Image Prior",
"url": "https://arxiv.org/abs/2601.08749",
"version": "v2"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "f90001dc-3345-45a9-a8d7-bdd089e4ac15",
"id": "arXiv Dataset",
"type": "Model",
"variant": "snapshot-2026-01-17",
"version": "0.1.0"
},
"user_id": 1000002
}