EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Wasserstein Optimal Transport Algorithms for High-Dimensional Image Reconstruction and Denoising

Suman Panwar

Boletim da Sociedade Paranaense de Matemática · 19.08.2026 · 44 (7) · 1-7

DOI: 10.5269/bspm.80587 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Boletim da Sociedade Paranaense de Matemática, online erschienen am 19.08.2026. Verfasst von Suman Panwar. Die Arbeit nennt 39 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

High-dimensional image reconstruction and denoising are still the major challenges of computational imaging in the modern era. Usually, traditional Euclidean fidelity measures and local smoothness priors fail to keep geometric structures and distributional relationships intact in natural images. Wasserstein Optimal Transport (OT) is a principled geometric framework for comparing probability distributions which has recently been used as a powerful instrument for image restoration, thus providing robustness to spatial rearrangements and better perceptual fidelity. We create a coherent theoretical and algorithmic framework for high-dimensional reconstruction and denoising using Wasserstein distances and their scalable approximations in this paper. Our method combines entropic OT through Sinkhorn iterations, projection-based and sliced Wasserstein formulations, and latent Gaussian OT mappings for deep feature distributions. We establish the rigorous mathematical basis such as existence and uniqueness of solutions, stability with respect to noise, and convergence guarantees for OT solvers. Extensive experiments validate that the proposed methods are superior to the classical variational techniques and contemporary deep baselines in structural preservation, texture consistency, and perceptual realism.

Zitieren

APA

Panwar, S. (2026). Wasserstein Optimal Transport Algorithms for High-Dimensional Image Reconstruction and Denoising. Boletim da Sociedade Paranaense de Matemática, 44(7), 1-7. https://doi.org/10.5269/bspm.80587

Vancouver

Panwar S. Wasserstein Optimal Transport Algorithms for High-Dimensional Image Reconstruction and Denoising. Boletim da Sociedade Paranaense de Matemática. 2026;44(7):1-7. doi:10.5269/bspm.80587

BibTeX
@article{Panwar2026,
  author = {Panwar, Suman},
  title = {Wasserstein Optimal Transport Algorithms for High-Dimensional Image Reconstruction and Denoising},
  journal = {Boletim da Sociedade Paranaense de Matemática},
  year = {2026},
  volume = {44},
  number = {7},
  pages = {1--7},
  doi = {10.5269/bspm.80587}
}
Kostenlos starten →