| bibtype |
J -
Journal Article
|
| ARLID |
0647611 |
| utime |
20260417124737.0 |
| mtime |
20260318235959.9 |
| SCOPUS |
105033249378 |
| WOS |
001719798300001 |
| DOI |
10.1080/23311916.2026.2641308 |
| title
(primary) (eng) |
Evaluating Texture Quality and Similarity Metrics for Visual Recognition and Modeling |
| specification |
| page_count |
25 s. |
| media_type |
E |
|
| serial |
| ARLID |
cav_un_epca*0634767 |
| ISSN |
2331-1916 |
| title
|
Cogent Engineering |
| volume_id |
13 |
|
| keyword |
Human quality ranking |
| keyword |
texture quality criteria |
| keyword |
image quality criteria |
| keyword |
texture quality benchmark |
| keyword |
Spearman correlation |
| author
(primary) |
| ARLID |
cav_un_auth*0101093 |
| name1 |
Haindl |
| name2 |
Michal |
| institution |
UTIA-B |
| full_dept (cz) |
Rozpoznávání obrazu |
| full_dept (eng) |
Department of Pattern Recognition |
| department (cz) |
RO |
| department (eng) |
RO |
| full_dept |
Department of Pattern Recognition |
| garant |
K |
| fullinstit |
Ústav teorie informace a automatizace AV ČR, v. v. i. |
|
| author
|
| ARLID |
cav_un_auth*0505664 |
| name1 |
Shaikh |
| name2 |
Nahidbanu |
| institution |
UTIA-B |
| full_dept (cz) |
Rozpoznávání obrazu |
| full_dept |
Department of Pattern Recognition |
| department (cz) |
RO |
| department |
RO |
| country |
IN |
| fullinstit |
Ústav teorie informace a automatizace AV ČR, v. v. i. |
|
| source |
|
| source |
|
| cas_special |
| abstract
(eng) |
Visual scene recognition and modeling fundamentally rely on visual textures, which encapsulate an object's varying material properties. The texture of a single material exhibits variations in scale illumination and viewing angles, influenced by factors such as the object's shape, alterations in lighting, and observational distance. This study aims to comprehensively compare various texture quality and similarity metrics proposed in prior research, evaluating their alignment with psychophysically derived human rankings using our texture quality benchmark. The ultimate goal is to replace cumbersome, costly, and impractical psychophysical evaluations-currently required for every synthesis or recognition experiment-with more efficient, objective numerical alternatives suitable for recognition and physically accurate texture modeling. The statistically analyzed quality measures either originate from texture quality assessment research or were initially developed for general image evaluation but can be applied to textures without requiring pixelwise correspondence between compared samples. Such numerical measures are fundamental for visual scene recognition and physically correct synthesis or editing applications in virtual prototyping. Our findings highlight the outstanding performance of two multispectral measures derived from the adaptive recursively evaluated Markovian descriptive model. Despite considerable progress in recent years, there remains room for enhancement in the current state-of-the-art texture quality assessment methods. The evaluated approaches require additional refinement to ensure their reliability across various computer vision and pattern recognition domains. |
| result_subspec |
WOS |
| RIV |
BD |
| FORD0 |
20000 |
| FORD1 |
20200 |
| FORD2 |
20205 |
| reportyear |
2027 |
| num_of_auth |
2 |
| inst_support |
RVO:67985556 |
| permalink |
https://hdl.handle.net/11104/0378167 |
| cooperation |
| ARLID |
cav_un_auth*0417618 |
| name |
Faculty of Information Technology, Czech Technical University in Prague |
| country |
NL |
|
| confidential |
S |
| article_num |
2641308 |
| mrcbC91 |
A |
| mrcbT16-e |
ENGINEERING.MULTIDISCIPLINARY |
| mrcbT16-f |
2.2 |
| mrcbT16-g |
0.7 |
| mrcbT16-h |
4.3 |
| mrcbT16-i |
0.00269 |
| mrcbT16-j |
0.324 |
| mrcbT16-k |
4023 |
| mrcbT16-q |
51 |
| mrcbT16-s |
0.476 |
| mrcbT16-y |
47.45 |
| mrcbT16-x |
3.51 |
| mrcbT16-3 |
2355 |
| mrcbT16-4 |
Q2 |
| mrcbT16-5 |
2.400 |
| mrcbT16-6 |
356 |
| mrcbT16-7 |
Q2 |
| mrcbT16-C |
72.3 |
| mrcbT16-M |
0.63 |
| mrcbT16-N |
Q2 |
| mrcbT16-P |
72.3 |
| arlyear |
2026 |
| mrcbU14 |
105033249378 SCOPUS |
| mrcbU24 |
PUBMED |
| mrcbU34 |
001719798300001 WOS |
| mrcbU63 |
cav_un_epca*0634767 Cogent Engineering 13 1 2026 2331-1916 2331-1916 |
|