Camera-independent color constancy by scene semantics
•Existing learning-based color constancy methods typically exhibit poor generalization between images taken by varying cameras.•Gray-based assumptions hold the natural property of camera-independent.•Gray-based assumptions are less accurate in different scenarios due to their fixity.•Scene semantics...
Saved in:
Published in | Pattern recognition letters Vol. 171; pp. 106 - 115 |
---|---|
Main Authors | , , , |
Format | Journal Article |
Language | English |
Published |
Elsevier B.V
01.07.2023
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Abstract | •Existing learning-based color constancy methods typically exhibit poor generalization between images taken by varying cameras.•Gray-based assumptions hold the natural property of camera-independent.•Gray-based assumptions are less accurate in different scenarios due to their fixity.•Scene semantics can contribute to inferring and modifying fixed gray-based assumptions.
Current learning-based color constancy methods are typically employed to find camera-specific illuminant mappings. Consequently, these methods exhibit poor generalization to images captured by varying cameras. In this paper, we present a camera-independent learning method based on Scene Semantics, and we call it CISS. Inspired by the camera-independent property of gray-based methods, CISS does not directly estimate camera-specific illuminant by training model as most learning methods do. Instead, the model's output is transformed into camera-independent scene statistics related to gray-based assumptions to avoid being affected by camera variations. Based on these estimated scene statistics, illuminant can be calculated indirectly. To estimate scene statistics accurately, CISS designs illuminant-invariance scene semantics features as input to the model. Then, the model estimates scene statistics for each input image in terms of scene semantics with exemplar-based learning. Experiments show that, on several public datasets, CISS is able to outperform present methods for multi-cameras color constancy, and is flexible enough to be well generalized to the unseen camera without fine-tuning by additional images. |
---|---|
AbstractList | •Existing learning-based color constancy methods typically exhibit poor generalization between images taken by varying cameras.•Gray-based assumptions hold the natural property of camera-independent.•Gray-based assumptions are less accurate in different scenarios due to their fixity.•Scene semantics can contribute to inferring and modifying fixed gray-based assumptions.
Current learning-based color constancy methods are typically employed to find camera-specific illuminant mappings. Consequently, these methods exhibit poor generalization to images captured by varying cameras. In this paper, we present a camera-independent learning method based on Scene Semantics, and we call it CISS. Inspired by the camera-independent property of gray-based methods, CISS does not directly estimate camera-specific illuminant by training model as most learning methods do. Instead, the model's output is transformed into camera-independent scene statistics related to gray-based assumptions to avoid being affected by camera variations. Based on these estimated scene statistics, illuminant can be calculated indirectly. To estimate scene statistics accurately, CISS designs illuminant-invariance scene semantics features as input to the model. Then, the model estimates scene statistics for each input image in terms of scene semantics with exemplar-based learning. Experiments show that, on several public datasets, CISS is able to outperform present methods for multi-cameras color constancy, and is flexible enough to be well generalized to the unseen camera without fine-tuning by additional images. |
Author | Xie, Mengda Lang, Yubo Sun, Peng Fang, Meie |
Author_xml | – sequence: 1 givenname: Mengda orcidid: 0000-0003-1068-1092 surname: Xie fullname: Xie, Mengda organization: Department of Public Security Information Technology and Intelligence, Criminal Investigation Police University of China, Shenyang 110035, China – sequence: 2 givenname: Peng surname: Sun fullname: Sun, Peng email: sunpeng_sx@cipuc.edu.cn organization: Department of Public Security Information Technology and Intelligence, Criminal Investigation Police University of China, Shenyang 110035, China – sequence: 3 givenname: Yubo surname: Lang fullname: Lang, Yubo organization: Department of Public Security Information Technology and Intelligence, Criminal Investigation Police University of China, Shenyang 110035, China – sequence: 4 givenname: Meie orcidid: 0000-0003-4292-8889 surname: Fang fullname: Fang, Meie organization: Department of Computer Science and Cyber Engineering, Guangzhou University, Guangzhou 511400, China |
BookMark | eNp9T81KAzEYDKJgW30DD_sCu-Z3s70IUvyDghc9hy9fvoWUNluSRejbm1LPwjBzmRlmluw6TYkYexC8E1z0j7vuCHMm7CSXquMV0l6xhRisbK3S-potqs22Q2_MLVuWsuOc92o9LJjZwIEytDEFOlKlNDc47adcOZUZEp4af2oKUqKm0AHSHLHcsZsR9oXu_3TFvl9fvjbv7fbz7WPzvG1RGjO3IDGMpq7QngsNWqhB41quSWFvgQ_C-CC91yNIj1YFUfePI2jwlgdJUq2YvvRinkrJNLpjjgfIJye4O193O3e57s7XHa-QtsaeLjGq234iZVcwUkIKsVpnF6b4f8Ev83xmKA |
Cites_doi | 10.1109/TIP.2007.901808 10.1109/TPAMI.2013.169 10.1038/scientificamerican1277-108 10.1007/s11263-015-0844-7 10.1364/OE.27.025611 10.1364/JOSAA.32.001707 10.1109/TIP.2011.2118224 10.1109/TIP.2017.2713044 10.1364/JOSAA.398692 10.1016/0016-0032(80)90058-7 10.1364/JOSAA.31.001049 10.1109/ACCESS.2021.3064382 10.1109/TPAMI.2010.93 |
ContentType | Journal Article |
Copyright | 2023 |
Copyright_xml | – notice: 2023 |
DBID | AAYXX CITATION |
DOI | 10.1016/j.patrec.2023.03.027 |
DatabaseName | CrossRef |
DatabaseTitle | CrossRef |
DatabaseTitleList | |
DeliveryMethod | fulltext_linktorsrc |
Discipline | Engineering Computer Science |
EISSN | 1872-7344 |
EndPage | 115 |
ExternalDocumentID | 10_1016_j_patrec_2023_03_027 S0167865523001010 |
GroupedDBID | --K --M .DC .~1 0R~ 123 1B1 1RT 1~. 1~5 29O 4.4 457 4G. 53G 5VS 7-5 71M 8P~ 9JN AABNK AACTN AAEDT AAEDW AAIAV AAIKJ AAKOC AALRI AAOAW AAQFI AAQXK AAXUO AAYFN ABBOA ABFNM ABFRF ABJNI ABMAC ABXDB ABYKQ ACDAQ ACGFO ACGFS ACNNM ACRLP ACZNC ADBBV ADEZE ADJOM ADMUD ADMXK ADTZH AEBSH AECPX AEFWE AEKER AENEX AFKWA AFTJW AGHFR AGUBO AGYEJ AHHHB AHJVU AHZHX AIALX AIEXJ AIKHN AITUG AJBFU AJOXV ALMA_UNASSIGNED_HOLDINGS AMFUW AMRAJ AOUOD ASPBG AVWKF AXJTR AZFZN BJAXD BKOJK BLXMC CS3 DU5 EBS EFJIC EFLBG EJD EO8 EO9 EP2 EP3 F5P FDB FEDTE FGOYB FIRID FNPLU FYGXN G-Q G8K GBLVA GBOLZ HLZ HVGLF HZ~ IHE J1W JJJVA KOM LG9 LY1 M41 MO0 N9A O-L O9- OAUVE OZT P-8 P-9 P2P PC. Q38 R2- RIG RNS ROL RPZ SBC SDF SDG SDP SDS SES SEW SPC SPCBC SST SSV SSZ T5K TN5 UNMZH VOH WH7 WUQ XFK XPP Y6R ZMT ~G- AAXKI AAYXX AFJKZ AKRWK CITATION |
ID | FETCH-LOGICAL-c255t-a2cdf58724b014a41384c929e3c67a0815bd2bb4fa2bc73d1202ffa4ab70d2e23 |
IEDL.DBID | .~1 |
ISSN | 0167-8655 |
IngestDate | Thu Sep 26 18:43:16 EDT 2024 Fri Feb 23 02:35:36 EST 2024 |
IsPeerReviewed | true |
IsScholarly | true |
Keywords | Illuminant estimation Gray-based assumptions Color constancy Scene semantics |
Language | English |
LinkModel | DirectLink |
MergedId | FETCHMERGED-LOGICAL-c255t-a2cdf58724b014a41384c929e3c67a0815bd2bb4fa2bc73d1202ffa4ab70d2e23 |
ORCID | 0000-0003-4292-8889 0000-0003-1068-1092 |
PageCount | 10 |
ParticipantIDs | crossref_primary_10_1016_j_patrec_2023_03_027 elsevier_sciencedirect_doi_10_1016_j_patrec_2023_03_027 |
PublicationCentury | 2000 |
PublicationDate | July 2023 2023-07-00 |
PublicationDateYYYYMMDD | 2023-07-01 |
PublicationDate_xml | – month: 07 year: 2023 text: July 2023 |
PublicationDecade | 2020 |
PublicationTitle | Pattern recognition letters |
PublicationYear | 2023 |
Publisher | Elsevier B.V |
Publisher_xml | – name: Elsevier B.V |
References | Cheng, Prasad, Brown (bib0026) 2014; 31 Gao, Zhang, Li (bib0029) 2019; 27 Hernandez-Juarez, Parisot, Busam (bib0015) 2020 Afifi, Brown (bib0019) 2019 Wu, Luo, Dang (bib0024) 2015; 32 Afifi, Barron, LeGendre (bib0013) 2021 Buchsbaum (bib0009) 1980; 310 McDonagh, S., Parisot, S., Zhou, F., et al., 2018. Formulating camera-adaptive color constancy as a few-shot meta-learning problem. arXiv preprint arXiv Xiao, Gu, Zhang (bib0017) 2020 Xu, Liu, Hou (bib0008) 2020 Gijsenij, Gevers, van de Weijer (bib0001) 2011; 20 Van de Weijer, Gevers, Gijsenij (bib0012) 2007; 16 Buzzelli, van de Weijer, Schettini (bib0032) 2018 Finlayson (bib0003) 2013 Gehler, Rother, Blake (bib0028) 2008 Li, Xiong, Hu (bib0020) 2016; 117 Buzzelli, Bianco, Schettini (bib0030) 2020; 37 Land (bib0010) 1977; 237 Nguyen, Prasad, Brown (bib0022) 2014 Gijsenij, Gevers (bib0014) 2011; 33 Lo, Chang, Chiu (bib0007) 2021 Finlayson, Trezzi (bib0011) 2004 Cheng, Price, Cohen (bib0005) 2015 Chakrabarti (bib0004) 2015 . Qian, Kamarainen, Nikkanen (bib0002) 2019 Barron, Tsai (bib0006) 2017 Wu (bib0025) 2010 Joze, Drew (bib0031) 2014; 36 Bianco, Cusano (bib0018) 2019 Laakom, Raitoharju, Nikkanen (bib0027) 2021; 9 Lazebnik, Schmid, Ponce (bib0023) 2006 Bianco, Cusano, Schettini (bib0021) 2017; 26 Lo (10.1016/j.patrec.2023.03.027_bib0007) 2021 Buchsbaum (10.1016/j.patrec.2023.03.027_bib0009) 1980; 310 Xu (10.1016/j.patrec.2023.03.027_bib0008) 2020 Joze (10.1016/j.patrec.2023.03.027_bib0031) 2014; 36 Gijsenij (10.1016/j.patrec.2023.03.027_bib0014) 2011; 33 Gao (10.1016/j.patrec.2023.03.027_bib0029) 2019; 27 Bianco (10.1016/j.patrec.2023.03.027_bib0021) 2017; 26 Bianco (10.1016/j.patrec.2023.03.027_bib0018) 2019 Van de Weijer (10.1016/j.patrec.2023.03.027_bib0012) 2007; 16 Finlayson (10.1016/j.patrec.2023.03.027_bib0003) 2013 Lazebnik (10.1016/j.patrec.2023.03.027_bib0023) 2006 Land (10.1016/j.patrec.2023.03.027_bib0010) 1977; 237 Xiao (10.1016/j.patrec.2023.03.027_bib0017) 2020 10.1016/j.patrec.2023.03.027_bib0016 Wu (10.1016/j.patrec.2023.03.027_bib0024) 2015; 32 Afifi (10.1016/j.patrec.2023.03.027_bib0013) 2021 Hernandez-Juarez (10.1016/j.patrec.2023.03.027_bib0015) 2020 Chakrabarti (10.1016/j.patrec.2023.03.027_bib0004) 2015 Afifi (10.1016/j.patrec.2023.03.027_bib0019) 2019 Wu (10.1016/j.patrec.2023.03.027_bib0025) 2010 Li (10.1016/j.patrec.2023.03.027_bib0020) 2016; 117 Cheng (10.1016/j.patrec.2023.03.027_bib0026) 2014; 31 Gehler (10.1016/j.patrec.2023.03.027_bib0028) 2008 Gijsenij (10.1016/j.patrec.2023.03.027_bib0001) 2011; 20 Barron (10.1016/j.patrec.2023.03.027_bib0006) 2017 Buzzelli (10.1016/j.patrec.2023.03.027_bib0032) 2018 Nguyen (10.1016/j.patrec.2023.03.027_bib0022) 2014 Laakom (10.1016/j.patrec.2023.03.027_bib0027) 2021; 9 Buzzelli (10.1016/j.patrec.2023.03.027_bib0030) 2020; 37 Qian (10.1016/j.patrec.2023.03.027_bib0002) 2019 Cheng (10.1016/j.patrec.2023.03.027_bib0005) 2015 Finlayson (10.1016/j.patrec.2023.03.027_bib0011) 2004 |
References_xml | – start-page: 8053 year: 2021 end-page: 8063 ident: bib0007 article-title: CLCC: contrastive learning for color constancy publication-title: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition contributor: fullname: Chiu – volume: 26 start-page: 4347 year: 2017 end-page: 4362 ident: bib0021 article-title: Single and multiple illuminant estimation using convolutional neural networks publication-title: IEEE Trans. Image Process contributor: fullname: Schettini – volume: 16 start-page: 2207 year: 2007 end-page: 2214 ident: bib0012 article-title: Edge-based color constancy publication-title: IEEE Trans. Image Process contributor: fullname: Gijsenij – start-page: 1 year: 2019 end-page: 8 ident: bib0019 article-title: Sensor-Independent Illumination Estimation for DNN Models publication-title: Proceedings of the British Machine Vision Conference contributor: fullname: Brown – volume: 117 start-page: 21 year: 2016 end-page: 47 ident: bib0020 article-title: Multi-cue illumination estimation publication-title: Int. J. Comput. Vis. contributor: fullname: Hu – volume: 33 start-page: 687 year: 2011 end-page: 698 ident: bib0014 article-title: Color constancy using natural image statistics and scene semantics publication-title: IEEE Trans. Pattern Anal. Mach. Intell. contributor: fullname: Gevers – start-page: 1904 year: 2013 end-page: 1911 ident: bib0003 article-title: Corrected-moment illuminant estimation publication-title: Proceedings of the IEEE International Conference on Computer Vision contributor: fullname: Finlayson – volume: 310 start-page: 1 year: 1980 end-page: 26 ident: bib0009 article-title: A spatial processor model for object colour perception publication-title: J. Frankl. Inst. contributor: fullname: Buchsbaum – start-page: 1 year: 2008 end-page: 8 ident: bib0028 article-title: Bayesian color constancy revisited publication-title: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition contributor: fullname: Blake – volume: 32 start-page: 1707 year: 2015 end-page: 1716 ident: bib0024 article-title: Edge-moment-based color constancy using illumination-coherent regularized regression publication-title: J. Opt. Soc. Am. A contributor: fullname: Dang – volume: 31 start-page: 1049 year: 2014 end-page: 1058 ident: bib0026 article-title: Illuminant estimation for color constancy: why spatial-domain methods work and the role of the color distribution publication-title: J. Opt. Soc. Am. A contributor: fullname: Brown – start-page: 2267 year: 2020 end-page: 2277 ident: bib0015 article-title: A multi-hypothesis approach to color constancy publication-title: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition contributor: fullname: Busam – start-page: 6950 year: 2017 end-page: 6958 ident: bib0006 article-title: Fast Fourier color constancy publication-title: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition contributor: fullname: Tsai – start-page: 552 year: 2010 end-page: 565 ident: bib0025 article-title: A fast dual method for HIK SVM learning publication-title: Proceedings of the European Conference on Computer Vision contributor: fullname: Wu – volume: 36 start-page: 860 year: 2014 end-page: 873 ident: bib0031 article-title: Exemplar-based color constancy and multiple illumination publication-title: IEEE Trans. Pattern Anal. Mach. Intell. contributor: fullname: Drew – start-page: 3398 year: 2014 end-page: 3405 ident: bib0022 article-title: Raw-to-raw: mapping between image sensor color responses publication-title: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition contributor: fullname: Brown – start-page: 2169 year: 2006 end-page: 2178 ident: bib0023 article-title: Beyond bags of features: spatial pyramid matching for recognizing natural scene categories publication-title: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition contributor: fullname: Ponce – start-page: 12204 year: 2019 end-page: 12213 ident: bib0018 article-title: Quasi-unsupervised color constancy publication-title: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition contributor: fullname: Cusano – volume: 20 start-page: 2475 year: 2011 end-page: 2489 ident: bib0001 article-title: Computational color constancy: survey and experiments publication-title: IEEE Trans. Image Process contributor: fullname: van de Weijer – volume: 27 start-page: 25611 year: 2019 end-page: 25633 ident: bib0029 article-title: Improving color constancy by selecting suitable set of training images publication-title: Opt. Express contributor: fullname: Li – start-page: 3255 year: 2020 end-page: 3264 ident: bib0017 article-title: Multi-domain learning for accurate and few-shot color constancy publication-title: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition contributor: fullname: Zhang – volume: 237 start-page: 108 year: 1977 end-page: 128 ident: bib0010 article-title: The retinex theory of color vision publication-title: Sci. Am. contributor: fullname: Land – start-page: 1000 year: 2015 end-page: 1008 ident: bib0005 article-title: Effective learning-based illuminant estimation using simple features publication-title: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition contributor: fullname: Cohen – volume: 37 start-page: 1721 year: 2020 end-page: 1730 ident: bib0030 article-title: ARC: angle-retaining chromaticity diagram for color constancy error analysis publication-title: J. Opt. Soc. Am. A contributor: fullname: Schettini – start-page: 8054 year: 2019 end-page: 8062 ident: bib0002 article-title: On finding gray pixels publication-title: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition contributor: fullname: Nikkanen – start-page: 3613 year: 2020 end-page: 3622 ident: bib0008 article-title: End-to-end illuminant estimation based on deep metric learning publication-title: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition contributor: fullname: Hou – start-page: 1981 year: 2021 end-page: 1990 ident: bib0013 article-title: Cross-camera convolutional color constancy publication-title: Proceedings of the IEEE International Conference on Computer Vision contributor: fullname: LeGendre – volume: 9 start-page: 39560 year: 2021 end-page: 39567 ident: bib0027 article-title: INTEL-TAU: a color constancy dataset publication-title: IEEE Access contributor: fullname: Nikkanen – start-page: 37 year: 2004 end-page: 41 ident: bib0011 article-title: Shades of gray and colour constancy publication-title: Proceedings of the IS&T/SID 12th Color Imaging Conference contributor: fullname: Trezzi – start-page: 3234 year: 2018 end-page: 3238 ident: bib0032 article-title: Learning illuminant estimation from object recognition publication-title: Proceedings of the International Conference on Image Processing contributor: fullname: Schettini – year: 2015 ident: bib0004 article-title: Color constancy by learning to predict chromaticity from luminance publication-title: Proceedings of the Advances in Neural Information Processing Systems contributor: fullname: Chakrabarti – start-page: 3234 year: 2018 ident: 10.1016/j.patrec.2023.03.027_bib0032 article-title: Learning illuminant estimation from object recognition contributor: fullname: Buzzelli – start-page: 3398 year: 2014 ident: 10.1016/j.patrec.2023.03.027_bib0022 article-title: Raw-to-raw: mapping between image sensor color responses contributor: fullname: Nguyen – volume: 16 start-page: 2207 year: 2007 ident: 10.1016/j.patrec.2023.03.027_bib0012 article-title: Edge-based color constancy publication-title: IEEE Trans. Image Process doi: 10.1109/TIP.2007.901808 contributor: fullname: Van de Weijer – volume: 36 start-page: 860 year: 2014 ident: 10.1016/j.patrec.2023.03.027_bib0031 article-title: Exemplar-based color constancy and multiple illumination publication-title: IEEE Trans. Pattern Anal. Mach. Intell. doi: 10.1109/TPAMI.2013.169 contributor: fullname: Joze – start-page: 1000 year: 2015 ident: 10.1016/j.patrec.2023.03.027_bib0005 article-title: Effective learning-based illuminant estimation using simple features contributor: fullname: Cheng – start-page: 8053 year: 2021 ident: 10.1016/j.patrec.2023.03.027_bib0007 article-title: CLCC: contrastive learning for color constancy contributor: fullname: Lo – start-page: 1904 year: 2013 ident: 10.1016/j.patrec.2023.03.027_bib0003 article-title: Corrected-moment illuminant estimation contributor: fullname: Finlayson – volume: 237 start-page: 108 year: 1977 ident: 10.1016/j.patrec.2023.03.027_bib0010 article-title: The retinex theory of color vision publication-title: Sci. Am. doi: 10.1038/scientificamerican1277-108 contributor: fullname: Land – start-page: 552 year: 2010 ident: 10.1016/j.patrec.2023.03.027_bib0025 article-title: A fast dual method for HIK SVM learning contributor: fullname: Wu – volume: 117 start-page: 21 year: 2016 ident: 10.1016/j.patrec.2023.03.027_bib0020 article-title: Multi-cue illumination estimation via a tree-structured group joint sparse representation publication-title: Int. J. Comput. Vis. doi: 10.1007/s11263-015-0844-7 contributor: fullname: Li – volume: 27 start-page: 25611 issue: 18 year: 2019 ident: 10.1016/j.patrec.2023.03.027_bib0029 article-title: Improving color constancy by selecting suitable set of training images publication-title: Opt. Express doi: 10.1364/OE.27.025611 contributor: fullname: Gao – start-page: 1 year: 2019 ident: 10.1016/j.patrec.2023.03.027_bib0019 article-title: Sensor-Independent Illumination Estimation for DNN Models publication-title: Proceedings of the British Machine Vision Conference contributor: fullname: Afifi – start-page: 1 year: 2008 ident: 10.1016/j.patrec.2023.03.027_bib0028 article-title: Bayesian color constancy revisited contributor: fullname: Gehler – volume: 32 start-page: 1707 year: 2015 ident: 10.1016/j.patrec.2023.03.027_bib0024 article-title: Edge-moment-based color constancy using illumination-coherent regularized regression publication-title: J. Opt. Soc. Am. A doi: 10.1364/JOSAA.32.001707 contributor: fullname: Wu – start-page: 3255 year: 2020 ident: 10.1016/j.patrec.2023.03.027_bib0017 article-title: Multi-domain learning for accurate and few-shot color constancy contributor: fullname: Xiao – start-page: 8054 year: 2019 ident: 10.1016/j.patrec.2023.03.027_bib0002 article-title: On finding gray pixels contributor: fullname: Qian – start-page: 3613 year: 2020 ident: 10.1016/j.patrec.2023.03.027_bib0008 article-title: End-to-end illuminant estimation based on deep metric learning contributor: fullname: Xu – start-page: 37 year: 2004 ident: 10.1016/j.patrec.2023.03.027_bib0011 article-title: Shades of gray and colour constancy contributor: fullname: Finlayson – ident: 10.1016/j.patrec.2023.03.027_bib0016 – volume: 20 start-page: 2475 year: 2011 ident: 10.1016/j.patrec.2023.03.027_bib0001 article-title: Computational color constancy: survey and experiments publication-title: IEEE Trans. Image Process doi: 10.1109/TIP.2011.2118224 contributor: fullname: Gijsenij – start-page: 6950 year: 2017 ident: 10.1016/j.patrec.2023.03.027_bib0006 article-title: Fast Fourier color constancy contributor: fullname: Barron – start-page: 2169 year: 2006 ident: 10.1016/j.patrec.2023.03.027_bib0023 article-title: Beyond bags of features: spatial pyramid matching for recognizing natural scene categories contributor: fullname: Lazebnik – start-page: 12204 year: 2019 ident: 10.1016/j.patrec.2023.03.027_bib0018 article-title: Quasi-unsupervised color constancy contributor: fullname: Bianco – volume: 26 start-page: 4347 year: 2017 ident: 10.1016/j.patrec.2023.03.027_bib0021 article-title: Single and multiple illuminant estimation using convolutional neural networks publication-title: IEEE Trans. Image Process doi: 10.1109/TIP.2017.2713044 contributor: fullname: Bianco – volume: 37 start-page: 1721 year: 2020 ident: 10.1016/j.patrec.2023.03.027_bib0030 article-title: ARC: angle-retaining chromaticity diagram for color constancy error analysis publication-title: J. Opt. Soc. Am. A doi: 10.1364/JOSAA.398692 contributor: fullname: Buzzelli – start-page: 1981 year: 2021 ident: 10.1016/j.patrec.2023.03.027_bib0013 article-title: Cross-camera convolutional color constancy contributor: fullname: Afifi – start-page: 2267 year: 2020 ident: 10.1016/j.patrec.2023.03.027_bib0015 article-title: A multi-hypothesis approach to color constancy contributor: fullname: Hernandez-Juarez – volume: 310 start-page: 1 year: 1980 ident: 10.1016/j.patrec.2023.03.027_bib0009 article-title: A spatial processor model for object colour perception publication-title: J. Frankl. Inst. doi: 10.1016/0016-0032(80)90058-7 contributor: fullname: Buchsbaum – volume: 31 start-page: 1049 year: 2014 ident: 10.1016/j.patrec.2023.03.027_bib0026 article-title: Illuminant estimation for color constancy: why spatial-domain methods work and the role of the color distribution publication-title: J. Opt. Soc. Am. A doi: 10.1364/JOSAA.31.001049 contributor: fullname: Cheng – volume: 9 start-page: 39560 year: 2021 ident: 10.1016/j.patrec.2023.03.027_bib0027 article-title: INTEL-TAU: a color constancy dataset publication-title: IEEE Access doi: 10.1109/ACCESS.2021.3064382 contributor: fullname: Laakom – volume: 33 start-page: 687 year: 2011 ident: 10.1016/j.patrec.2023.03.027_bib0014 article-title: Color constancy using natural image statistics and scene semantics publication-title: IEEE Trans. Pattern Anal. Mach. Intell. doi: 10.1109/TPAMI.2010.93 contributor: fullname: Gijsenij – year: 2015 ident: 10.1016/j.patrec.2023.03.027_bib0004 article-title: Color constancy by learning to predict chromaticity from luminance contributor: fullname: Chakrabarti |
SSID | ssj0006398 |
Score | 2.4529517 |
Snippet | •Existing learning-based color constancy methods typically exhibit poor generalization between images taken by varying cameras.•Gray-based assumptions hold the... |
SourceID | crossref elsevier |
SourceType | Aggregation Database Publisher |
StartPage | 106 |
SubjectTerms | Color constancy Gray-based assumptions Illuminant estimation Scene semantics |
Title | Camera-independent color constancy by scene semantics |
URI | https://dx.doi.org/10.1016/j.patrec.2023.03.027 |
Volume | 171 |
hasFullText | 1 |
inHoldings | 1 |
isFullTextHit | |
isPrint | |
link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnV1LS8NAEB5KvejBR1Wsj5KD17Vxd5NsjqVYqmIvWugt7GY3EMFa2vTQi7_dmTy0gngQQiBLFsLMZr6ZYeYbgGsnLLGi-cwXKmTyFgMUdEs0s0parcNMxJZSA0-TcDyVD7Ng1oJh0wtDZZW17a9semmt65V-Lc3-Is_7z1RAT22V6EQTURrF7RLBCM_0zcd3mQcisGr4ventpn2urPGifLMjIkMuSqpTmi3zGzxtQc7oEPZrX9EbVJ9zBC0378BBM4fBq3_LDuxtkQoeQzDUlGhi-deA28Ijauol3ktfMN14ZuMRi5PzVu4NRZunqxOYju5ehmNWD0dgKUYBBdM8tVmgIi4NRjkasUjJFH0dJ9Iw0gj0gbHcGJlpbtJIWNQCzzIttYl8yx0Xp9Cev8_dGXixDiSqRSJchjJUNo6EjnVshMoUt5Z3gTUySRYVB0bSFIe9JpUME5Jh4uPFoy5EjeCSH7pM0Ez_ufP83zsvYJeeqkLaS2gXy7W7QnehML3yPPRgZ3D_OJ58AqBrvto |
link.rule.ids | 315,786,790,4521,24144,27957,27958,45620,45714 |
linkProvider | Elsevier |
linkToHtml | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnV1LS8NAEB6KHtSDj6pYnzl4XRt3N6-jFEvVthdb6G3ZzW6ggrW08dCLv92ZPLSCeBBCDkkWwkwy3zfD7DcA105YUkXzmS_ikMlbTFCQlmhmY2m1DjORWCoNDIZhbywfJ8GkAZ16Lwy1VVaxv4zpRbSurrQra7bn02n7mRroaVslkmgSSsO8fZPoPM1vuPn47vNACI5rgW96vN4_VzR5UcHZkZIhF4XWKQ2X-Q2f1jCnuw-7FVn07sr3OYCGmzVhrx7E4FX_ZRN21lQFDyHoaKo0senXhNvcI23qBZ4LMpiuPLPySMbJeUv3iradpssjGHfvR50eq6YjsBTTgJxpntosiCMuDaY5GsEolimSHSfSMNKI9IGx3BiZaW7SSFh0A88yLbWJfMsdF8ewMXubuRPwEh1I9ItEvAxlGNskEjrRiRFxFnNreQtYbRM1L0UwVN0d9qJKGyqyofLx4FELotpw6oczFcbpP1ee_nvlFWz1RoO-6j8Mn85gm-6UXbXnsJEv3t0FcofcXBbfxidINcBs |
openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Camera-independent+color+constancy+by+scene+semantics&rft.jtitle=Pattern+recognition+letters&rft.au=Xie%2C+Mengda&rft.au=Sun%2C+Peng&rft.au=Lang%2C+Yubo&rft.au=Fang%2C+Meie&rft.date=2023-07-01&rft.issn=0167-8655&rft.volume=171&rft.spage=106&rft.epage=115&rft_id=info:doi/10.1016%2Fj.patrec.2023.03.027&rft.externalDBID=n%2Fa&rft.externalDocID=10_1016_j_patrec_2023_03_027 |
thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0167-8655&client=summon |
thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0167-8655&client=summon |
thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0167-8655&client=summon |