Multi-resolution deconvolution
Mirza Karamehmedović, Pierre Maréchal, Martin Sæbye Carøe, and Lara Baalbaki · 2025
Read Multi-resolution deconvolution by Mirza Karamehmedović, Pierre Maréchal, Martin Sæbye Carøe, and Lara Baalbaki (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2512-23342.txt.
Description
Mirza Karamehmedović, Pierre Maréchal, Martin Carøe, and Lara Baalbaki generalize deconvolution, the art of undoing a blur. Classical deconvolution assumes one fixed blurring kernel everywhere, but many imaging problems blur unevenly across the scene. Their framework allows a pseudodifferential solution operator whose symbol varies with both position and frequency, which translates into images whose sharpness can vary across space while keeping one global stability promise.
The theory also allows fairly general convolution kernels, widening the class of forward models covered. Consistency, convergence, and stability results are proved, together with convergence rates, so the new method comes with the same kind of backing as the classical one. Examples back the analysis and show the advantages of letting sharpness adapt across space. The authors position the framework for imaging tasks where a single uniform sharpness is the wrong goal.
Catalog record
| CREATOR | Mirza Karamehmedović, Pierre Maréchal, Martin Sæbye Carøe, and Lara Baalbaki |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 33 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2512-23342