Guided Variational Network for Image Decomposition
SIAM Journal on Imaging Sciences · 31.12.2026 · 19 (4) · 2581-2614
DOI: 10.1137/26m1841203 ↗Kurz gefasst
Fachartikel in der Zeitschrift SIAM Journal on Imaging Sciences, online erschienen am 18.11.2026. Verfasst von Alessandro Lanza und 3 weiteren Autor:innen. Die Arbeit nennt 36 Literaturangaben. Gefördert von: PNRR CN-HPC, INdAM–GNCS, National Natural Science Foundation of China.
Abstract (Originalsprache)
Cartoon-texture image decomposition is a critical preprocessing problem bottlenecked by the numerical intractability of classical variational or optimization models and the tedious manual tuning of global regularization parameters. We propose a Guided Variational Decomposition (GVD) model which introduces spatially adaptive quadratic norms whose pixelwise weights are learned either through local probabilistic statistics or via a lightweight neural network within a bilevel framework. This leads to a unified, interpretable, and computationally efficient model that bridges classical variational ideas with modern adaptive and data-driven methodologies. Numerical experiments on this framework, which inherently includes automatic parameter selection, delivers GVD as a robust, self-tuning, and superior solution for reliable image decomposition.
Förderung
- PNRR CN-HPC
- INdAM–GNCS
- National Natural Science Foundation of China
- Postgraduate Scientific Research Innovation Project of Hunan Province
Zitieren
APA
Lanza, A., Morigi, S., Wen, Y., & Yang, L. (2026). Guided Variational Network for Image Decomposition. SIAM Journal on Imaging Sciences, 19(4), 2581-2614. https://doi.org/10.1137/26m1841203
Vancouver
Lanza A, Morigi S, Wen Y, Yang L. Guided Variational Network for Image Decomposition. SIAM Journal on Imaging Sciences. 2026;19(4):2581-2614. doi:10.1137/26m1841203
BibTeX
@article{Lanza2026,
author = {Lanza, Alessandro and Morigi, Serena and Wen, Youwei and Yang, Li},
title = {Guided Variational Network for Image Decomposition},
journal = {SIAM Journal on Imaging Sciences},
year = {2026},
volume = {19},
number = {4},
pages = {2581--2614},
doi = {10.1137/26m1841203}
}
EUVIMED