| bibtype |
C -
Conference Paper (international conference)
|
| ARLID |
0583748 |
| utime |
20240429140505.4 |
| mtime |
20240306235959.9 |
| DOI |
10.5220/0012397600003660 |
| title
(primary) (eng) |
Avoiding Undesirable Solutions of Deep Blind Image Deconvolution |
| specification |
| page_count |
7 s. |
| media_type |
E |
|
| serial |
| ARLID |
cav_un_epca*0583747 |
| ISBN |
978-989-758-679-8 |
| ISSN |
2184-4321 |
| title
|
Proceedings of the 19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2024) |
| page_num |
559-566 |
| publisher |
| place |
Setúbal |
| name |
SciTePress |
| year |
2024 |
|
| editor |
|
| editor |
| name1 |
Furnari |
| name2 |
Antonino |
|
| editor |
| name1 |
Bouatouch |
| name2 |
Kadi |
|
| editor |
| name1 |
Sousa |
| name2 |
A. Augusto |
|
|
| keyword |
Blind Image Deconvolution |
| keyword |
Deep Image Prior |
| keyword |
No-Blur |
| keyword |
Variational Bayes |
| author
(primary) |
| ARLID |
cav_un_auth*0464277 |
| name1 |
Brožová |
| name2 |
Antonie |
| institution |
UTIA-B |
| full_dept (cz) |
Adaptivní systémy |
| full_dept (eng) |
Department of Adaptive Systems |
| department (cz) |
AS |
| department (eng) |
AS |
| country |
CZ |
| fullinstit |
Ústav teorie informace a automatizace AV ČR, v. v. i. |
|
| author
|
| ARLID |
cav_un_auth*0101207 |
| name1 |
Šmídl |
| name2 |
Václav |
| institution |
UTIA-B |
| department |
AS |
| full_dept (cz) |
Adaptivní systémy |
| full_dept |
Department of Adaptive Systems |
| department (cz) |
AS |
| department |
AS |
| fullinstit |
Ústav teorie informace a automatizace AV ČR, v. v. i. |
|
| source |
|
| cas_special |
| project |
| project_id |
GA20-27939S |
| agency |
GA ČR |
| ARLID |
cav_un_auth*0391986 |
|
| project |
| project_id |
GA24-10400S |
| agency |
GA ČR |
| country |
CZ |
| ARLID |
cav_un_auth*0464279 |
|
| abstract
(eng) |
Blind image deconvolution (BID) is a severely ill-posed optimization problem requiring additional information, typically in the form of regularization. Deep image prior (DIP) promises to model a naturally looking image due to a well-chosen structure of a neural network. The use of DIP in BID results in a significant perfor-mance improvement in terms of average PSNR. In this contribution, we offer qualitative analysis of selected DIP-based methods w.r.t. two types of undesired solutions: blurred image (no-blur) and a visually corrupted image (solution with artifacts). We perform a sensitivity study showing which aspects of the DIP-based algorithms help to avoid which undesired mode. We confirm that the no-blur can be avoided using either sharp image prior or tuning of the hyperparameters of the optimizer. The artifact solution is a harder problem since variations that suppress the artifacts often suppress good solutions as well. Switching to the structural similarity index measure fro m L 2 norm in loss was found to be the most successful approach to mitigate the artifacts. |
| action |
| ARLID |
cav_un_auth*0464278 |
| name |
International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2024) /19./ |
| dates |
20240227 |
| mrcbC20-s |
20240229 |
| place |
Roma |
| country |
IT |
|
| RIV |
IN |
| FORD0 |
10000 |
| FORD1 |
10200 |
| FORD2 |
10201 |
| reportyear |
2025 |
| num_of_auth |
2 |
| presentation_type |
PR |
| inst_support |
RVO:67985556 |
| permalink |
https://hdl.handle.net/11104/0353240 |
| confidential |
S |
| arlyear |
2024 |
| mrcbU14 |
SCOPUS |
| mrcbU24 |
PUBMED |
| mrcbU34 |
WOS |
| mrcbU63 |
cav_un_epca*0583747 Proceedings of the 19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2024) SciTePress 2024 Setúbal 559 566 978-989-758-679-8 2184-4321 |
| mrcbU67 |
Radeva Petia 340 |
| mrcbU67 |
Furnari Antonino 340 |
| mrcbU67 |
Bouatouch Kadi 340 |
| mrcbU67 |
Sousa A. Augusto 340 |
|