Bilinear Convolutional Neural Networks for Fine-Grained Visual Recognition
We present a simple and effective architecture for fine-grained recognition called Bilinear Convolutional Neural Networks (B-CNNs). These networks represent an image as a pooled outer product of features derived from two CNNs and capture localized feature interactions in a translationally invariant...
Uloženo v:
| Vydáno v: | IEEE transactions on pattern analysis and machine intelligence Ročník 40; číslo 6; s. 1309 - 1322 |
|---|---|
| Hlavní autoři: | , , |
| Médium: | Journal Article |
| Jazyk: | angličtina |
| Vydáno: |
United States
IEEE
01.06.2018
The Institute of Electrical and Electronics Engineers, Inc. (IEEE) |
| Témata: | |
| ISSN: | 0162-8828, 1939-3539, 2160-9292, 1939-3539 |
| On-line přístup: | Získat plný text |
| Tagy: |
Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
|
| Abstract | We present a simple and effective architecture for fine-grained recognition called Bilinear Convolutional Neural Networks (B-CNNs). These networks represent an image as a pooled outer product of features derived from two CNNs and capture localized feature interactions in a translationally invariant manner. B-CNNs are related to orderless texture representations built on deep features but can be trained in an end-to-end manner. Our most accurate model obtains 84.1, 79.4, 84.5 and 91.3 percent per-image accuracy on the Caltech-UCSD birds [1], NABirds [2], FGVC aircraft [3], and Stanford cars [4] dataset respectively and runs at 30 frames-persecond on a NVIDIA Titan X GPU. We then present a systematic analysis of these networks and show that (1) the bilinear features are highly redundant and can be reduced by an order of magnitude in size without significant loss in accuracy, (2) are also effective for other image classification tasks such as texture and scene recognition, and (3) can be trained from scratch on the ImageNet dataset offering consistent improvements over the baseline architecture. Finally, we present visualizations of these models on various datasets using top activations of neural units and gradient-based inversion techniques. The source code for the complete system is available at http://vis-www.cs.umass.edu/bcnn. |
|---|---|
| AbstractList | We present a simple and effective architecture for fine-grained recognition called Bilinear Convolutional Neural Networks (B-CNNs). These networks represent an image as a pooled outer product of features derived from two CNNs and capture localized feature interactions in a translationally invariant manner. B-CNNs are related to orderless texture representations built on deep features but can be trained in an end-to-end manner. Our most accurate model obtains 84.1, 79.4, 84.5 and 91.3 percent per-image accuracy on the Caltech-UCSD birds [1], NABirds [2], FGVC aircraft [3], and Stanford cars [4] dataset respectively and runs at 30 frames-persecond on a NVIDIA Titan X GPU. We then present a systematic analysis of these networks and show that (1) the bilinear features are highly redundant and can be reduced by an order of magnitude in size without significant loss in accuracy, (2) are also effective for other image classification tasks such as texture and scene recognition, and (3) can be trained from scratch on the ImageNet dataset offering consistent improvements over the baseline architecture. Finally, we present visualizations of these models on various datasets using top activations of neural units and gradient-based inversion techniques. The source code for the complete system is available at http://vis-www.cs.umass.edu/bcnn. We present a simple and effective architecture for fine-grained recognition called Bilinear Convolutional Neural Networks (B-CNNs). These networks represent an image as a pooled outer product of features derived from two CNNs and capture localized feature interactions in a translationally invariant manner. B-CNNs are related to orderless texture representations built on deep features but can be trained in an end-to-end manner. Our most accurate model obtains 84.1%, 79.4%, 84.5% and 91.3% per-image accuracy on the Caltech-UCSD birds [66], NABirds [63], FGVC aircraft [42], and Stanford cars [33] dataset respectively and runs at 30 frames-per-second on a NVIDIA Titan X GPU. We then present a systematic analysis of these networks and show that (1) the bilinear features are highly redundant and can be reduced by an order of magnitude in size without significant loss in accuracy, (2) are also effective for other image classification tasks such as texture and scene recognition, and (3) can be trained from scratch on the ImageNet dataset offering consistent improvements over the baseline architecture. Finally, we present visualizations of these models on various datasets using top activations of neural units and gradient-based inversion techniques. The source code for the complete system is available at http://vis-www.cs.umass.edu/bcnn. We present a simple and effective architecture for fine-grained recognition called Bilinear Convolutional Neural Networks (B-CNNs). These networks represent an image as a pooled outer product of features derived from two CNNs and capture localized feature interactions in a translationally invariant manner. B-CNNs are related to orderless texture representations built on deep features but can be trained in an end-to-end manner. Our most accurate model obtains 84.1%, 79.4%, 84.5% and 91.3% per-image accuracy on the Caltech-UCSD birds [66], NABirds [63], FGVC aircraft [42], and Stanford cars [33] dataset respectively and runs at 30 frames-per-second on a NVIDIA Titan X GPU. We then present a systematic analysis of these networks and show that (1) the bilinear features are highly redundant and can be reduced by an order of magnitude in size without significant loss in accuracy, (2) are also effective for other image classification tasks such as texture and scene recognition, and (3) can be trained from scratch on the ImageNet dataset offering consistent improvements over the baseline architecture. Finally, we present visualizations of these models on various datasets using top activations of neural units and gradient-based inversion techniques. The source code for the complete system is available at http://vis-www.cs.umass.edu/bcnn.We present a simple and effective architecture for fine-grained recognition called Bilinear Convolutional Neural Networks (B-CNNs). These networks represent an image as a pooled outer product of features derived from two CNNs and capture localized feature interactions in a translationally invariant manner. B-CNNs are related to orderless texture representations built on deep features but can be trained in an end-to-end manner. Our most accurate model obtains 84.1%, 79.4%, 84.5% and 91.3% per-image accuracy on the Caltech-UCSD birds [66], NABirds [63], FGVC aircraft [42], and Stanford cars [33] dataset respectively and runs at 30 frames-per-second on a NVIDIA Titan X GPU. We then present a systematic analysis of these networks and show that (1) the bilinear features are highly redundant and can be reduced by an order of magnitude in size without significant loss in accuracy, (2) are also effective for other image classification tasks such as texture and scene recognition, and (3) can be trained from scratch on the ImageNet dataset offering consistent improvements over the baseline architecture. Finally, we present visualizations of these models on various datasets using top activations of neural units and gradient-based inversion techniques. The source code for the complete system is available at http://vis-www.cs.umass.edu/bcnn. |
| Author | RoyChowdhury, Aruni Maji, Subhransu Tsung-Yu Lin |
| Author_xml | – sequence: 1 surname: Tsung-Yu Lin fullname: Tsung-Yu Lin email: tsungyulin@cs.umass.edu organization: Coll. of Inf. & Comput. Sci., Univ. of Massachusetts Amherst, Amherst, MA, USA – sequence: 2 givenname: Aruni surname: RoyChowdhury fullname: RoyChowdhury, Aruni email: arunirc@cs.umass.edu organization: Coll. of Inf. & Comput. Sci., Univ. of Massachusetts Amherst, Amherst, MA, USA – sequence: 3 givenname: Subhransu surname: Maji fullname: Maji, Subhransu email: smaji@cs.umass.edu organization: Coll. of Inf. & Comput. Sci., Univ. of Massachusetts Amherst, Amherst, MA, USA |
| BackLink | https://www.ncbi.nlm.nih.gov/pubmed/28692962$$D View this record in MEDLINE/PubMed |
| BookMark | eNp9kU1P4zAQhi3ECgrLHwAJReLCJV2Pncb2ESo-itgPIeBqOc4EGdIY7ATEv8el7R44cHovzzOamXeHbHa-Q0L2gY4BqPp1--_k92zMKIgxE4wXlG6QEYOS5ooptklGFEqWS8nkNtmJ8ZFSKCaUb5FtJsuElGxErk5d6zo0IZv67tW3Q-98Z9rsDw7hM_o3H55i1viQnScwvwgmRZ3duzgk4Aatf-jcwvpJfjSmjbi3yl1yd352O73Mr_9ezKYn17nlnPV5OeGiTPujBcugxspyRChVVWBhOFU1CsuskLLhRWNAMMUrWVcNVgIqCch3yfFy7nPwLwPGXs9dtNi2pkM_RA0KhConUkBCj76gj34I6b6oGYiiUFRKmqjDFTVUc6z1c3BzE971-ksJkEvABh9jwEZb15vFzX36RquB6kUh-rMQvShErwpJKvuirqd_Kx0sJYeI_4V0lOQT4B-kmJWK |
| CODEN | ITPIDJ |
| CitedBy_id | crossref_primary_10_1007_s00521_024_10861_4 crossref_primary_10_1109_ACCESS_2019_2953957 crossref_primary_10_1109_JSTARS_2022_3168691 crossref_primary_10_1109_TCYB_2020_2979262 crossref_primary_10_1109_JTEHM_2023_3241613 crossref_primary_10_1109_TAI_2023_3339593 crossref_primary_10_1109_TIP_2021_3094744 crossref_primary_10_1088_1742_6596_1544_1_012138 crossref_primary_10_1007_s11042_020_08666_3 crossref_primary_10_1007_s10489_022_04257_x crossref_primary_10_3390_s23083970 crossref_primary_10_1016_j_neucom_2020_02_101 crossref_primary_10_1109_TPAMI_2020_2974833 crossref_primary_10_1016_j_patcog_2021_108219 crossref_primary_10_1007_s10489_021_02925_y crossref_primary_10_1016_j_ins_2025_121970 crossref_primary_10_1109_TPAMI_2019_2933510 crossref_primary_10_1109_TIP_2019_2908795 crossref_primary_10_1109_TIP_2019_2934576 crossref_primary_10_1109_TIP_2021_3120319 crossref_primary_10_1109_TCSVT_2023_3284405 crossref_primary_10_1109_LGRS_2021_3070145 crossref_primary_10_1109_ACCESS_2019_2925002 crossref_primary_10_1109_TMM_2024_3366404 crossref_primary_10_1109_ACCESS_2018_2844789 crossref_primary_10_1109_TIP_2021_3089942 crossref_primary_10_3390_s21124176 crossref_primary_10_1007_s41095_022_0332_2 crossref_primary_10_1016_j_patcog_2023_109776 crossref_primary_10_1109_JBHI_2023_3299321 crossref_primary_10_1007_s11263_018_1125_z crossref_primary_10_1109_TGRS_2021_3079918 crossref_primary_10_3390_app15116085 crossref_primary_10_1007_s11063_023_11383_1 crossref_primary_10_1016_j_compbiomed_2023_106765 crossref_primary_10_1109_ACCESS_2021_3131002 crossref_primary_10_1007_s00521_023_09349_4 crossref_primary_10_1088_1742_6596_2050_1_012006 crossref_primary_10_3390_app9020301 crossref_primary_10_3390_app122412715 crossref_primary_10_1016_j_compag_2021_106679 crossref_primary_10_1109_ACCESS_2020_3030249 crossref_primary_10_1016_j_cviu_2022_103436 crossref_primary_10_1016_j_bbe_2023_12_003 crossref_primary_10_4018_JCIT_326131 crossref_primary_10_1016_j_eswa_2022_117674 crossref_primary_10_1016_j_patcog_2019_107167 crossref_primary_10_1007_s00371_020_02052_8 crossref_primary_10_1016_j_bspc_2022_103874 crossref_primary_10_1016_j_procs_2022_09_053 crossref_primary_10_3390_e26010029 crossref_primary_10_1088_1361_6560_ac0f30 crossref_primary_10_1080_10916466_2025_2468299 crossref_primary_10_1007_s00371_022_02686_w crossref_primary_10_1109_TII_2024_3435430 crossref_primary_10_3390_electronics14142869 crossref_primary_10_1016_j_asoc_2023_109997 crossref_primary_10_1109_TMM_2023_3244340 crossref_primary_10_1109_ACCESS_2019_2926994 crossref_primary_10_3390_e23070816 crossref_primary_10_1109_TII_2020_2985159 crossref_primary_10_23919_JSEE_2022_000155 crossref_primary_10_3390_jimaging8070186 crossref_primary_10_1109_ACCESS_2023_3292340 crossref_primary_10_1002_hbm_25685 crossref_primary_10_1007_s10409_021_09057_z crossref_primary_10_1016_j_compag_2022_107440 crossref_primary_10_1007_s00330_025_11385_8 crossref_primary_10_1016_j_neucom_2022_03_032 crossref_primary_10_1016_j_patcog_2020_107593 crossref_primary_10_1109_TIE_2019_2962437 crossref_primary_10_1016_j_neucom_2023_126652 crossref_primary_10_1109_TCSVT_2022_3216905 crossref_primary_10_1002_ima_22628 crossref_primary_10_1109_TII_2019_2908211 crossref_primary_10_1016_j_jvcir_2021_103368 crossref_primary_10_1109_TIE_2021_3086714 crossref_primary_10_1016_j_jvcir_2021_103084 crossref_primary_10_3390_jimaging11080283 crossref_primary_10_1109_TIP_2022_3193763 crossref_primary_10_1145_3489142 crossref_primary_10_1186_s13321_024_00844_x crossref_primary_10_1145_3447866 crossref_primary_10_1109_TPAMI_2018_2885764 crossref_primary_10_1155_2022_5375449 crossref_primary_10_1007_s10489_022_03232_w crossref_primary_10_1109_TMM_2020_3001510 crossref_primary_10_1016_j_eswa_2022_118223 crossref_primary_10_1007_s11042_022_13811_1 crossref_primary_10_1109_JSTARS_2020_3006241 crossref_primary_10_1007_s11042_022_12481_3 crossref_primary_10_1016_j_neucom_2022_04_037 crossref_primary_10_1016_j_asoc_2023_110477 crossref_primary_10_1109_ACCESS_2019_2898028 crossref_primary_10_1007_s11760_024_03693_1 crossref_primary_10_1109_TNNLS_2020_3044078 crossref_primary_10_1109_TNNLS_2021_3052829 crossref_primary_10_1109_TIP_2020_3019185 crossref_primary_10_1186_s13640_023_00613_0 crossref_primary_10_1109_TGRS_2023_3311093 crossref_primary_10_1109_TIM_2021_3139685 crossref_primary_10_3390_app15168903 crossref_primary_10_1109_TVT_2020_3030018 crossref_primary_10_1016_j_neucom_2020_01_067 crossref_primary_10_3390_app11199204 crossref_primary_10_1007_s11042_020_09679_8 crossref_primary_10_1016_j_imavis_2024_104923 crossref_primary_10_1016_j_patcog_2024_110749 crossref_primary_10_1109_ACCESS_2020_2986771 crossref_primary_10_3390_rs14061445 crossref_primary_10_3390_s22249650 crossref_primary_10_1016_j_compbiomed_2022_106424 crossref_primary_10_1016_j_ijleo_2018_11_145 crossref_primary_10_1109_TCSVT_2020_3020079 crossref_primary_10_1109_ACCESS_2019_2936118 crossref_primary_10_1109_TIP_2023_3283065 crossref_primary_10_1007_s11431_020_1777_4 crossref_primary_10_1007_s11042_021_11769_0 crossref_primary_10_1007_s11554_022_01228_w crossref_primary_10_1007_s11432_023_3922_2 crossref_primary_10_1016_j_atmosres_2021_105839 crossref_primary_10_1109_TIP_2022_3181492 crossref_primary_10_1016_j_patcog_2023_109979 crossref_primary_10_1109_TMM_2024_3369968 crossref_primary_10_1016_j_artmed_2021_102130 crossref_primary_10_1109_LSP_2023_3262059 crossref_primary_10_1002_jbio_202100370 crossref_primary_10_1007_s11432_020_3181_9 crossref_primary_10_3390_ani12213000 crossref_primary_10_1109_JSEN_2024_3415078 crossref_primary_10_1109_TCSVT_2019_2920783 crossref_primary_10_1109_TPAMI_2021_3107164 crossref_primary_10_1109_TCSVT_2024_3406443 crossref_primary_10_1109_TCYB_2024_3424430 crossref_primary_10_1109_TIM_2024_3522672 crossref_primary_10_1109_TPAMI_2022_3164894 crossref_primary_10_1007_s00521_024_10464_z crossref_primary_10_1109_TNNLS_2019_2947789 crossref_primary_10_1007_s11633_023_1451_7 crossref_primary_10_1016_j_neucom_2020_07_147 crossref_primary_10_1109_TAI_2023_3329457 crossref_primary_10_1109_TIP_2020_3032029 crossref_primary_10_1007_s12539_021_00479_8 crossref_primary_10_1016_j_bspc_2022_104007 crossref_primary_10_3390_app132111716 crossref_primary_10_1007_s11263_020_01389_w crossref_primary_10_1016_j_ins_2020_10_014 crossref_primary_10_1109_TITS_2019_2921732 crossref_primary_10_3390_rs15245759 crossref_primary_10_3390_rs14184467 crossref_primary_10_1109_LGRS_2019_2953754 crossref_primary_10_1016_j_patcog_2023_109550 crossref_primary_10_1109_TITS_2023_3274777 crossref_primary_10_1038_s41598_023_48916_6 crossref_primary_10_1109_TMM_2022_3158001 crossref_primary_10_1016_j_eswa_2020_113993 crossref_primary_10_1145_3570166 crossref_primary_10_1016_j_neucom_2019_06_013 crossref_primary_10_1109_TIP_2019_2924811 crossref_primary_10_1145_3300938 crossref_primary_10_1109_ACCESS_2018_2874994 crossref_primary_10_3390_w15050845 crossref_primary_10_1109_TMM_2022_3214431 crossref_primary_10_1109_TPAMI_2021_3126648 crossref_primary_10_1007_s11263_020_01376_1 crossref_primary_10_1007_s11668_023_01695_8 crossref_primary_10_1088_1361_6501_ad8592 crossref_primary_10_1109_TIP_2024_3404854 crossref_primary_10_1016_j_comcom_2021_03_006 crossref_primary_10_1109_TIP_2021_3049334 crossref_primary_10_1109_TPAMI_2019_2932058 crossref_primary_10_1109_JSEN_2024_3470250 crossref_primary_10_1109_TPAMI_2022_3215702 crossref_primary_10_1109_ACCESS_2020_2975913 crossref_primary_10_1016_j_imavis_2025_105548 crossref_primary_10_1109_JSEN_2023_3248868 crossref_primary_10_1016_j_patcog_2023_109439 crossref_primary_10_1109_OJIM_2022_3190024 crossref_primary_10_1109_TGRS_2024_3458408 crossref_primary_10_3389_fnins_2021_728874 crossref_primary_10_3390_app11146533 crossref_primary_10_3390_app15084155 crossref_primary_10_1007_s11042_023_15005_9 crossref_primary_10_1016_j_engappai_2022_105291 |
| Cites_doi | 10.1016/j.patrec.2014.06.011 10.1109/CVPR.2010.5540039 10.1109/CVPR.2015.7299035 10.1007/s11263-015-0816-y 10.1109/ICCV.2005.54 10.1109/ICCVW.2013.77 10.1109/CVPR.2016.90 10.1109/ICCV.2011.6126238 10.1007/978-3-642-33786-4_32 10.1007/978-3-319-10584-0_26 10.1007/978-3-319-46487-9_19 10.1109/ICCV.2011.6126413 10.1109/CVPRW.2014.131 10.18653/v1/D16-1044 10.1162/089976600300015349 10.1023/A:1011126920638 10.1109/ICCV.2015.170 10.1109/CVPR.2014.461 10.1109/CVPR.2016.41 10.1109/ICCV.1999.790410 10.1109/CVPR.2015.7298658 10.1145/2733373.2807412 10.1109/CVPR.2016.213 10.1007/978-3-319-10590-1_54 10.1109/CVPR.2016.128 10.1109/CVPR.2014.81 10.1145/1873951.1874249 10.1109/CVPR.2016.129 10.1023/A:1026553619983 10.1145/2487575.2487591 10.1109/CVPR.2016.305 10.1007/978-3-642-15561-1_11 10.1109/ICCV.2015.339 10.1109/CVPR.2016.572 10.1007/s11263-016-0911-8 10.5244/C.30.24 10.1109/CVPR.2015.7299194 10.1109/CVPR.2014.219 10.7551/mitpress/4175.001.0001 10.5244/C.28.6 10.1007/s11263-009-0275-4 10.1109/CVPR.2016.131 10.1109/CVPR.2012.6248364 10.1109/CVPR.2007.383266 10.1007/s11263-015-0872-3 10.1109/ICCV.2015.136 10.1109/CVPR.2016.265 10.1109/CVPR.2009.5206537 10.1109/CVPR.2015.7298594 |
| ContentType | Journal Article |
| Copyright | Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2018 |
| Copyright_xml | – notice: Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2018 |
| DBID | 97E RIA RIE AAYXX CITATION NPM 7SC 7SP 8FD JQ2 L7M L~C L~D 7X8 |
| DOI | 10.1109/TPAMI.2017.2723400 |
| DatabaseName | IEEE All-Society Periodicals Package (ASPP) 2005-present IEEE All-Society Periodicals Package (ASPP) 1998-Present IEEE Electronic Library (IEL) CrossRef PubMed Computer and Information Systems Abstracts Electronics & Communications Abstracts Technology Research Database ProQuest Computer Science Collection Advanced Technologies Database with Aerospace Computer and Information Systems Abstracts Academic Computer and Information Systems Abstracts Professional MEDLINE - Academic |
| DatabaseTitle | CrossRef PubMed Technology Research Database Computer and Information Systems Abstracts – Academic Electronics & Communications Abstracts ProQuest Computer Science Collection Computer and Information Systems Abstracts Advanced Technologies Database with Aerospace Computer and Information Systems Abstracts Professional MEDLINE - Academic |
| DatabaseTitleList | PubMed Technology Research Database MEDLINE - Academic |
| Database_xml | – sequence: 1 dbid: NPM name: PubMed url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed sourceTypes: Index Database – sequence: 2 dbid: RIE name: IEEE Electronic Library (IEL) url: https://ieeexplore.ieee.org/ sourceTypes: Publisher – sequence: 3 dbid: 7X8 name: MEDLINE - Academic url: https://search.proquest.com/medline sourceTypes: Aggregation Database |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Engineering Computer Science |
| EISSN | 2160-9292 1939-3539 |
| EndPage | 1322 |
| ExternalDocumentID | 28692962 10_1109_TPAMI_2017_2723400 7968351 |
| Genre | orig-research Research Support, U.S. Gov't, Non-P.H.S Research Support, Non-U.S. Gov't Journal Article |
| GrantInformation_xml | – fundername: Director of National Intelligence (ODNI) – fundername: Intelligence Advanced Research Projects Activity (IARPA) grantid: 2014-14071600010 – fundername: National Science Foundation grantid: IIS-1617917 |
| GroupedDBID | --- -DZ -~X .DC 0R~ 29I 4.4 53G 5GY 6IK 97E AAJGR AARMG AASAJ AAWTH ABAZT ABQJQ ABVLG ACGFO ACGFS ACIWK ACNCT AENEX AGQYO AHBIQ AKJIK AKQYR ALMA_UNASSIGNED_HOLDINGS ASUFR ATWAV BEFXN BFFAM BGNUA BKEBE BPEOZ CS3 DU5 E.L EBS EJD F5P HZ~ IEDLZ IFIPE IPLJI JAVBF LAI M43 MS~ O9- OCL P2P PQQKQ RIA RIE RNS RXW TAE TN5 UHB ~02 AAYXX CITATION 5VS 9M8 AAYOK ABFSI ADRHT AETEA AETIX AGSQL AI. AIBXA ALLEH FA8 H~9 IBMZZ ICLAB IFJZH NPM RIG RNI RZB VH1 XJT 7SC 7SP 8FD JQ2 L7M L~C L~D 7X8 |
| ID | FETCH-LOGICAL-c332t-65376110ec1c21debc3ee169b4e4a309de7c2c788f34fa17293b8dbfeb71b81e3 |
| IEDL.DBID | RIE |
| ISSN | 0162-8828 1939-3539 |
| IngestDate | Sun Sep 28 01:13:54 EDT 2025 Sun Jun 29 16:12:05 EDT 2025 Thu Apr 03 07:08:47 EDT 2025 Sat Nov 29 07:53:49 EST 2025 Tue Nov 18 21:32:39 EST 2025 Wed Aug 27 02:47:51 EDT 2025 |
| IsPeerReviewed | true |
| IsScholarly | true |
| Issue | 6 |
| Language | English |
| License | https://ieeexplore.ieee.org/Xplorehelp/downloads/license-information/IEEE.html https://doi.org/10.15223/policy-029 https://doi.org/10.15223/policy-037 |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-c332t-65376110ec1c21debc3ee169b4e4a309de7c2c788f34fa17293b8dbfeb71b81e3 |
| Notes | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14 content type line 23 |
| ORCID | 0000-0002-1332-646X |
| PMID | 28692962 |
| PQID | 2174490880 |
| PQPubID | 85458 |
| PageCount | 14 |
| ParticipantIDs | pubmed_primary_28692962 crossref_primary_10_1109_TPAMI_2017_2723400 ieee_primary_7968351 proquest_miscellaneous_1917965871 proquest_journals_2174490880 crossref_citationtrail_10_1109_TPAMI_2017_2723400 |
| PublicationCentury | 2000 |
| PublicationDate | 2018-06-01 |
| PublicationDateYYYYMMDD | 2018-06-01 |
| PublicationDate_xml | – month: 06 year: 2018 text: 2018-06-01 day: 01 |
| PublicationDecade | 2010 |
| PublicationPlace | United States |
| PublicationPlace_xml | – name: United States – name: New York |
| PublicationTitle | IEEE transactions on pattern analysis and machine intelligence |
| PublicationTitleAbbrev | TPAMI |
| PublicationTitleAlternate | IEEE Trans Pattern Anal Mach Intell |
| PublicationYear | 2018 |
| Publisher | IEEE The Institute of Electrical and Electronics Engineers, Inc. (IEEE) |
| Publisher_xml | – name: IEEE – name: The Institute of Electrical and Electronics Engineers, Inc. (IEEE) |
| References | ref57 ref13 ref12 (ref69) 2017 ref59 ref15 ref58 ref14 ref53 ref52 ref11 krizhevsky (ref17) 2012 (ref5) 2017 ref54 ref10 ioffe (ref68) 2015 gens (ref47) 2012 ref16 ref19 maji (ref3) 2013 liu (ref41) 2016 ref51 ref48 scholkopf (ref46) 2001 ref42 branson (ref24) 2014 ref44 jaderberg (ref32) 2015 zhang (ref66) 2016 ref8 ref7 ref9 ref4 ref40 simonyan (ref50) 2014 mnih (ref33) 2014 ref35 donahue (ref18) 2013 ref37 ref36 ref31 ref30 ref2 ref39 ref38 ren (ref55) 2015 simonyan (ref29) 2015 liu (ref56) 2016 sharan (ref70) 2009; 9 pirsiavash (ref49) 2009 ref71 wah (ref1) 2011 ustyuzhaninov (ref45) 2016 ref72 ref67 ref23 ref26 ref25 ref64 ref20 ref63 ref22 ref65 ref21 csurka (ref6) 2004 ref28 ref27 gatys (ref43) 2015 ba (ref34) 2015 ref60 ref62 ref61 |
| References_xml | – ident: ref27 doi: 10.1016/j.patrec.2014.06.011 – ident: ref59 doi: 10.1109/CVPR.2010.5540039 – volume: 9 year: 2009 ident: ref70 article-title: Material perceprion: What can you see in a brief glance? publication-title: J Vis – start-page: 21 year: 2016 ident: ref56 article-title: SSD: Single shot multibox detector publication-title: Proc Eur Conf Comput Vis – ident: ref51 doi: 10.1109/CVPR.2015.7299035 – start-page: 1482 year: 2009 ident: ref49 article-title: Bilinear classifiers for visual recognition publication-title: Proc Advances Neural Inf Process Syst – ident: ref7 doi: 10.1007/s11263-015-0816-y – year: 2015 ident: ref34 article-title: Multiple object recognition with visual attention publication-title: Proc Int Conf Learn Representations – ident: ref71 doi: 10.1109/ICCV.2005.54 – ident: ref4 doi: 10.1109/ICCVW.2013.77 – year: 2016 ident: ref45 article-title: Texture synthesis using shallow convolutional networks with random filters publication-title: arXiv preprint arXiv 1606 00021 – ident: ref31 doi: 10.1109/CVPR.2016.90 – ident: ref21 doi: 10.1109/ICCV.2011.6126238 – start-page: 3 year: 2014 ident: ref24 article-title: Bird species categorization using pose normalized deep convolutional nets publication-title: Proc British Mach Vis Conf – ident: ref11 doi: 10.1007/978-3-642-33786-4_32 – year: 2013 ident: ref3 article-title: Fine-grained visual classification of aircraft publication-title: arXiv preprint arXiv 1306 5151 – ident: ref10 doi: 10.1007/978-3-319-10584-0_26 – ident: ref35 doi: 10.1007/978-3-319-46487-9_19 – ident: ref20 doi: 10.1109/ICCV.2011.6126413 – ident: ref19 doi: 10.1109/CVPRW.2014.131 – start-page: 91 year: 2015 ident: ref55 article-title: Faster R-CNN: Towards real-time object detection with region proposal networks publication-title: Proc Advances Neural Inf Process Syst – year: 2011 ident: ref1 article-title: The Caltech-UCSD birds-200-2011 dataset – year: 2017 ident: ref69 – ident: ref52 doi: 10.18653/v1/D16-1044 – ident: ref48 doi: 10.1162/089976600300015349 – start-page: 1 year: 2004 ident: ref6 article-title: Visual categorization with bags of keypoints publication-title: Proc ECCV Workshop Statist Learn Comput Vis – start-page: 1097 year: 2012 ident: ref17 article-title: ImageNet classification with deep convolutional neural networks publication-title: Proc Advances Neural Inf Process Syst – start-page: 448 year: 2015 ident: ref68 article-title: Batch normalization: Accelerating deep network training by reducing internal covariate shift publication-title: Proc Int Conf Mach Learn – start-page: 2017 year: 2015 ident: ref32 article-title: Spatial transformer networks publication-title: Proc Advances Neural Inf Process Syst – ident: ref37 doi: 10.1023/A:1011126920638 – ident: ref13 doi: 10.1109/ICCV.2015.170 – ident: ref57 doi: 10.1109/CVPR.2014.461 – ident: ref15 doi: 10.1109/CVPR.2016.41 – ident: ref8 doi: 10.1109/ICCV.1999.790410 – ident: ref2 doi: 10.1109/CVPR.2015.7298658 – ident: ref12 doi: 10.1145/2733373.2807412 – ident: ref53 doi: 10.1109/CVPR.2016.213 – ident: ref23 doi: 10.1007/978-3-319-10590-1_54 – ident: ref64 doi: 10.1109/CVPR.2016.128 – ident: ref25 doi: 10.1109/CVPR.2014.81 – start-page: 3248 year: 2012 ident: ref47 article-title: Discriminative learning of sum-product networks publication-title: Proc Advances Neural Inf Process Syst – ident: ref62 doi: 10.1145/1873951.1874249 – ident: ref65 doi: 10.1109/CVPR.2016.129 – ident: ref38 doi: 10.1023/A:1026553619983 – ident: ref54 doi: 10.1145/2487575.2487591 – ident: ref14 doi: 10.1109/CVPR.2016.305 – ident: ref58 doi: 10.1007/978-3-642-15561-1_11 – start-page: 2204 year: 2014 ident: ref33 article-title: Recurrent models of visual attention publication-title: Proc Advances Neural Inf Process Syst – ident: ref61 doi: 10.1109/ICCV.2015.339 – ident: ref42 doi: 10.1109/CVPR.2016.572 – ident: ref16 doi: 10.1007/s11263-016-0911-8 – ident: ref36 doi: 10.5244/C.30.24 – ident: ref26 doi: 10.1109/CVPR.2015.7299194 – start-page: 568 year: 2014 ident: ref50 article-title: Two-stream convolutional networks for action recognition in videos publication-title: Proc Advances Neural Inf Process Syst – ident: ref60 doi: 10.1109/CVPR.2014.219 – year: 2001 ident: ref46 publication-title: Learning With Kernels Support Vector Machines Regularization Optimization and Beyond doi: 10.7551/mitpress/4175.001.0001 – ident: ref28 doi: 10.5244/C.28.6 – year: 2016 ident: ref66 article-title: Fine-grained pose prediction, normalization, and recognition publication-title: Proc Workshop Int Conf Learn Represent – ident: ref40 doi: 10.1007/s11263-009-0275-4 – ident: ref67 doi: 10.1109/CVPR.2016.131 – ident: ref22 doi: 10.1109/CVPR.2012.6248364 – ident: ref39 doi: 10.1109/CVPR.2007.383266 – start-page: 262 year: 2015 ident: ref43 article-title: Texture synthesis and the controlled generation of natural stimuli using convolutional neural networks publication-title: Proc Advances Neural Inf Process Syst – ident: ref9 doi: 10.1007/s11263-015-0872-3 – ident: ref63 doi: 10.1109/ICCV.2015.136 – year: 2013 ident: ref18 article-title: Decaf: A deep convolutional activation feature for generic visual recognition publication-title: Proc Int Conf Mach Learn – year: 2017 ident: ref5 – year: 2016 ident: ref41 article-title: Cross-convolutional-layer pooling for image recognition publication-title: IEEE Trans Pattern Anal Mach Intell – ident: ref44 doi: 10.1109/CVPR.2016.265 – ident: ref72 doi: 10.1109/CVPR.2009.5206537 – ident: ref30 doi: 10.1109/CVPR.2015.7298594 – year: 2015 ident: ref29 article-title: Very deep convolutional networks for large-scale image recognition publication-title: Proc Int Conf Learn Representations |
| SSID | ssj0014503 |
| Score | 2.6647253 |
| Snippet | We present a simple and effective architecture for fine-grained recognition called Bilinear Convolutional Neural Networks (B-CNNs). These networks represent an... |
| SourceID | proquest pubmed crossref ieee |
| SourceType | Aggregation Database Index Database Enrichment Source Publisher |
| StartPage | 1309 |
| SubjectTerms | Artificial neural networks bilinear models Birds Computer architecture Convolutional codes convolutional networks Feature extraction Fine-grained recognition Image classification Image recognition Neural networks Object recognition second order pooling Source code Texture recognition texture representations Visualization |
| Title | Bilinear Convolutional Neural Networks for Fine-Grained Visual Recognition |
| URI | https://ieeexplore.ieee.org/document/7968351 https://www.ncbi.nlm.nih.gov/pubmed/28692962 https://www.proquest.com/docview/2174490880 https://www.proquest.com/docview/1917965871 |
| Volume | 40 |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| journalDatabaseRights | – providerCode: PRVIEE databaseName: IEEE Electronic Library (IEL) customDbUrl: eissn: 2160-9292 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0014503 issn: 0162-8828 databaseCode: RIE dateStart: 19790101 isFulltext: true titleUrlDefault: https://ieeexplore.ieee.org/ providerName: IEEE |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LS8QwEB5W8aAH17frY6ngTaubpm3ao4rrAxURlb2VJp3AgrSy3fX3O0kfKKjgqYVO0pCZdL5pJt8AHJKLDXSQpa7QYej6zKfvYBAoV0fkOzIepAPLxPR6Jx4eotEofuzAcXsWBhFt8hmemFu7l58VamZ-lZ2KOCTAQLHOnBCiOqvV7hj4ga2CTAiGVjiFEc0BmUF8-vx4dn9jsrjEiSc8TlZrKICjkJBB6H3zR7bAyu9Y0_qcYfd_o12B5RpbOmeVMaxCB_M16DZ1G5x6Ga_B0hcSwnW4PR8bqJlOnIsi_6gtkboxtB32YvPES4fQrTMkQffKVJXAzHkdlzMSeGpSkIp8A16Gl88X125dYcFVnHtTNzRkLjRHqJjyWIZScUQWxtJHP-WDOEOhPEVRsua-TgnrxFxGmdQoBZMRQ74J83mR4zY4GGOqtUTy-cyXTKcyEzJAgldCGAaYHrBmnhNV04-bKhhviQ1DBnFi1ZQYNSW1mnpw1LZ5r8g3_pReN0poJev578Feo86kXp9lYgIxm-JFrQ7ax7SyzHZJmmMxKxMTyRpqHEFdbFVm0PbdWM_Oz-_chUUaWVSllO3B_HQyw31YUB_TcTnpk_mOor4130-xWOhr |
| linkProvider | IEEE |
| linkToHtml | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3db9MwED9NBWnjgcE-oFAgSHsbaevYiZPHMVE26Kpp6qa9WbFzliqhBDXt_n7OzodAGkg8JVLOjuU7534Xn38HcEIuNrZxkYfSJkkomKDvYByb0KbkOwoe51PPxHQ3l4tFen-fXe_Ap_4sDCL65DMcu1u_l19UZut-lU1klhBgoFjnSSxExJrTWv2egYh9HWTCMLTGKZDojshMs8ny-uzq0uVxyXEkI05260iA04SwQRL94ZF8iZW_o03vdWb7_zfeF_C8RZfBWWMOL2EHywPY7yo3BO1CPoBnv9EQHsK3zysHNvN1cF6VD60tUjeOuMNffKZ4HRC-DWYkGH51dSWwCO5W9ZYEbrokpKo8gtvZl-X5RdjWWAgN59EmTBydC80RGmYiVqA2HJElmRYocj7NCpQmMhQnWy5sTmgn4zottEUtmU4Z8mMYlFWJryHADHNrNZLXZ0Izm-tC6hgJYEnpOGCGwLp5VqYlIHd1MH4oH4hMM-XVpJyaVKumIZz2bX429Bv_lD50Sugl2_kfwqhTp2pXaK1cKOaTvKjVx_4xrS23YZKXWG1r5WJZR44jqYtXjRn0fXfW8-bxd36A3Yvl1VzNLxff38IejTJtEsxGMNist_gOnpqHzapev_dG_Aty_-rK |
| 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=Bilinear+Convolutional+Neural+Networks+for+Fine-grained+Visual+Recognition&rft.jtitle=IEEE+transactions+on+pattern+analysis+and+machine+intelligence&rft.au=Lin%2C+Tsung-Yu&rft.au=RoyChowdhury%2C+Aruni&rft.au=Maji%2C+Subhransu&rft.date=2018-06-01&rft.eissn=1939-3539&rft.volume=40&rft.issue=6&rft.spage=1309&rft_id=info:doi/10.1109%2FTPAMI.2017.2723400&rft_id=info%3Apmid%2F28692962&rft.externalDocID=28692962 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0162-8828&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0162-8828&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0162-8828&client=summon |