| bibtype |
J -
Journal Article
|
| ARLID |
0616904 |
| utime |
20250320140311.7 |
| mtime |
20250213235959.9 |
| SCOPUS |
85216108770 |
| WOS |
001420733900001 |
| DOI |
10.1016/j.dsp.2025.105022 |
| title
(primary) (eng) |
Implicit neural representation for image demosaicking |
| specification |
| page_count |
14 s. |
| media_type |
P |
|
| serial |
| ARLID |
cav_un_epca*0252719 |
| ISSN |
1051-2004 |
| title
|
Digital Signal Processing |
| volume_id |
159 |
| publisher |
|
|
| keyword |
Demosaicking |
| keyword |
Implicit neural representation |
| keyword |
Inverse problems |
| author
(primary) |
| ARLID |
cav_un_auth*0379363 |
| name1 |
Kerepecký |
| name2 |
Tomáš |
| institution |
UTIA-B |
| full_dept (cz) |
Zpracování obrazové informace |
| full_dept (eng) |
Department of Image Processing |
| department (cz) |
ZOI |
| department (eng) |
ZOI |
| full_dept |
Department of Image Processing |
| country |
CZ |
| garant |
K |
| fullinstit |
Ústav teorie informace a automatizace AV ČR, v. v. i. |
|
| author
|
| ARLID |
cav_un_auth*0101209 |
| name1 |
Šroubek |
| name2 |
Filip |
| institution |
UTIA-B |
| full_dept (cz) |
Zpracování obrazové informace |
| full_dept |
Department of Image Processing |
| department (cz) |
ZOI |
| department |
ZOI |
| full_dept |
Department of Image Processing |
| fullinstit |
Ústav teorie informace a automatizace AV ČR, v. v. i. |
|
| author
|
| ARLID |
cav_un_auth*0101087 |
| name1 |
Flusser |
| name2 |
Jan |
| institution |
UTIA-B |
| full_dept (cz) |
Zpracování obrazové informace |
| full_dept |
Department of Image Processing |
| department (cz) |
ZOI |
| department |
ZOI |
| full_dept |
Department of Image Processing |
| fullinstit |
Ústav teorie informace a automatizace AV ČR, v. v. i. |
|
| source |
|
| source |
|
| cas_special |
| project |
| project_id |
GA25-15933S |
| agency |
GA ČR |
| ARLID |
cav_un_auth*0483946 |
|
| abstract
(eng) |
We propose a novel approach to enhance image demosaicking algorithms using implicit neural representations (INR). Our method employs a multi-layer perceptron to encode RGB images, combining original Bayer measurements with an initial estimate from existing demosaicking methods to achieve superior reconstructions. A key innovation is the integration of two loss functions: a Bayer loss for fidelity to sensor data and a complementary loss that regularizes reconstruction using interpolated data from the initial estimate. This combination, along with INR’s inherent ability to capture fine details, enables hig-fidelity reconstructions that incorporate information from both sources. Furthermore, we demonstrate that INR can effectively correct artifacts in state-of-the-art demosaicking methods when input data diverge from the training distribution, such as in cases of noise or blur. This adaptability highlights the transformative potential of INR-based demosaicking, offering a robust solution to this challenging problem. |
| reportyear |
2026 |
| RIV |
IN |
| result_subspec |
WOS |
| FORD0 |
10000 |
| FORD1 |
10200 |
| FORD2 |
10201 |
| num_of_auth |
3 |
| inst_support |
RVO:67985556 |
| permalink |
https://hdl.handle.net/11104/0364269 |
| cooperation |
| ARLID |
cav_un_auth*0329918 |
| name |
FJFI ČVUT Praha |
| country |
CZ |
|
| confidential |
S |
| article_num |
105022 |
| mrcbC91 |
C |
| mrcbT16-e |
ENGINEERING.ELECTRICAL&ELECTRONIC |
| mrcbT16-f |
2.9 |
| mrcbT16-g |
0.6 |
| mrcbT16-h |
4.3 |
| mrcbT16-i |
0.00715 |
| mrcbT16-j |
0.589 |
| mrcbT16-k |
7998 |
| mrcbT16-q |
96 |
| mrcbT16-s |
0.704 |
| mrcbT16-y |
45.4 |
| mrcbT16-x |
3.83 |
| mrcbT16-3 |
4297 |
| mrcbT16-4 |
Q2 |
| mrcbT16-5 |
2.600 |
| mrcbT16-6 |
492 |
| mrcbT16-7 |
Q2 |
| mrcbT16-C |
58 |
| mrcbT16-M |
0.67 |
| mrcbT16-N |
Q2 |
| mrcbT16-P |
57.5 |
| arlyear |
2025 |
| mrcbU14 |
85216108770 SCOPUS |
| mrcbU24 |
PUBMED |
| mrcbU34 |
001420733900001 WOS |
| mrcbU63 |
cav_un_epca*0252719 Digital Signal Processing 159 1 2025 1051-2004 1095-4333 Elsevier |
|