View-Invariant Deep Architecture for Human Action Recognition Using Two-Stream Motion and Shape Temporal Dynamics

Human action Recognition for unknown views, is a challenging task. We propose a deep view-invariant human action recognition framework, which is a novel integration of two important action cues: motion and shape temporal dynamics (STD). The motion stream encapsulates the motion content of action as...

Full description

Saved in:

Bibliographic Details
Published in	IEEE transactions on image processing Vol. 29; pp. 3835 - 3844
Main Authors	Dhiman, Chhavi, Vishwakarma, Dinesh Kumar
Format	Journal Article
Language	English
Published	United States IEEE 01.01.2020 The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Subjects	Algorithms Computer architecture Datasets Dynamics Human action recognition Human activity recognition Human motion human pose model Image recognition Invariants late fusion Shape Shape recognition spatial temporal dynamics Streaming media Trajectory Videos
Online Access	Get full text

Cover

Loading…

Abstract	Human action Recognition for unknown views, is a challenging task. We propose a deep view-invariant human action recognition framework, which is a novel integration of two important action cues: motion and shape temporal dynamics (STD). The motion stream encapsulates the motion content of action as RGB Dynamic Images (RGB-DIs), which are generated by Approximate Rank Pooling (ARP) and processed by using fine-tuned InceptionV3 model. The STD stream learns long-term view-invariant shape dynamics of action using a sequence of LSTM and Bi-LSTM learning models. Human Pose Model (HPM) generates view-invariant features of structural similarity index matrix (SSIM) based key depth human pose frames. The final prediction of the action is made on the basis of three types of late fusion techniques i.e. maximum (max), average (avg) and multiply (mul), applied on individual stream scores. To validate the performance of the proposed novel framework, the experiments are performed using both cross-subject and cross-view validation schemes on three publically available benchmarks-NUCLA multi-view dataset, UWA3D-II Activity dataset and NTU RGB-D Activity dataset. Our algorithm outperforms existing state-of-the-arts significantly, which is measured in terms of recognition accuracy, receiver operating characteristic (ROC) curve and area under the curve (AUC).
AbstractList	Human action Recognition for unknown views, is a challenging task. We propose a deep view-invariant human action recognition framework, which is a novel integration of two important action cues: motion and shape temporal dynamics (STD). The motion stream encapsulates the motion content of action as RGB Dynamic Images (RGB-DIs), which are generated by Approximate Rank Pooling (ARP) and processed by using finetuned InceptionV3 model. The STD stream learns long-term view-invariant shape dynamics of action using a sequence of LSTM and Bi-LSTM learning models. Human Pose Model (HPM) generates view-invariant features of structural similarity index matrix (SSIM) based key depth human pose frames. The final prediction of the action is made on the basis of three types of late fusion techniques i.e. maximum (max), average (avg) and multiply (mul), applied on individual stream scores. To validate the performance of the proposed novel framework, the experiments are performed using both cross-subject and cross-view validation schemes on three publically available benchmarks- NUCLA multi-view dataset, UWA3D-II Activity dataset and NTU RGB-D Activity dataset. Our algorithm outperforms existing state-of-the-arts significantly, which is measured in terms of recognition accuracy, receiver operating characteristic (ROC) curve and area under the curve (AUC). Human action Recognition for unknown views, is a challenging task. We propose a deep view-invariant human action recognition framework, which is a novel integration of two important action cues: motion and shape temporal dynamics (STD). The motion stream encapsulates the motion content of action as RGB Dynamic Images (RGB-DIs), which are generated by Approximate Rank Pooling (ARP) and processed by using fine-tuned InceptionV3 model. The STD stream learns long-term view-invariant shape dynamics of action using a sequence of LSTM and Bi-LSTM learning models. Human Pose Model (HPM) generates view-invariant features of structural similarity index matrix (SSIM) based key depth human pose frames. The final prediction of the action is made on the basis of three types of late fusion techniques i.e. maximum (max), average (avg) and multiply (mul), applied on individual stream scores. To validate the performance of the proposed novel framework, the experiments are performed using both cross-subject and cross-view validation schemes on three publically available benchmarks-NUCLA multi-view dataset, UWA3D-II Activity dataset and NTU RGB-D Activity dataset. Our algorithm outperforms existing state-of-the-arts significantly, which is measured in terms of recognition accuracy, receiver operating characteristic (ROC) curve and area under the curve (AUC). Human action Recognition for unknown views, is a challenging task. We propose a deep view-invariant human action recognition framework, which is a novel integration of two important action cues: motion and shape temporal dynamics (STD). The motion stream encapsulates the motion content of action as RGB Dynamic Images (RGB-DIs), which are generated by Approximate Rank Pooling (ARP) and processed by using finetuned InceptionV3 model. The STD stream learns long-term view-invariant shape dynamics of action using a sequence of LSTM and Bi-LSTM learning models. Human Pose Model (HPM) generates view-invariant features of structural similarity index matrix (SSIM) based key depth human pose frames. The final prediction of the action is made on the basis of three types of late fusion techniques i.e. maximum (max), average (avg) and multiply (mul), applied on individual stream scores. To validate the performance of the proposed novel framework, the experiments are performed using both cross-subject and cross-view validation schemes on three publically available benchmarks- NUCLA multi-view dataset, UWA3D-II Activity dataset and NTU RGB-D Activity dataset. Our algorithm outperforms existing state-of-the-arts significantly, which is measured in terms of recognition accuracy, receiver operating characteristic (ROC) curve and area under the curve (AUC).Human action Recognition for unknown views, is a challenging task. We propose a deep view-invariant human action recognition framework, which is a novel integration of two important action cues: motion and shape temporal dynamics (STD). The motion stream encapsulates the motion content of action as RGB Dynamic Images (RGB-DIs), which are generated by Approximate Rank Pooling (ARP) and processed by using finetuned InceptionV3 model. The STD stream learns long-term view-invariant shape dynamics of action using a sequence of LSTM and Bi-LSTM learning models. Human Pose Model (HPM) generates view-invariant features of structural similarity index matrix (SSIM) based key depth human pose frames. The final prediction of the action is made on the basis of three types of late fusion techniques i.e. maximum (max), average (avg) and multiply (mul), applied on individual stream scores. To validate the performance of the proposed novel framework, the experiments are performed using both cross-subject and cross-view validation schemes on three publically available benchmarks- NUCLA multi-view dataset, UWA3D-II Activity dataset and NTU RGB-D Activity dataset. Our algorithm outperforms existing state-of-the-arts significantly, which is measured in terms of recognition accuracy, receiver operating characteristic (ROC) curve and area under the curve (AUC).
Author	Vishwakarma, Dinesh Kumar Dhiman, Chhavi
Author_xml	– sequence: 1 givenname: Chhavi surname: Dhiman fullname: Dhiman, Chhavi organization: Department of Information Technology, Biometric Research Laboratory, Delhi Technological University (formerly Delhi College of Engineering), New Delhi, India – sequence: 2 givenname: Dinesh Kumar orcidid: 0000-0002-1026-0047 surname: Vishwakarma fullname: Vishwakarma, Dinesh Kumar email: dvishwakarma@gmail.com organization: Department of Information Technology, Biometric Research Laboratory, Delhi Technological University (formerly Delhi College of Engineering), New Delhi, India
BackLink	https://www.ncbi.nlm.nih.gov/pubmed/31944975$$D View this record in MEDLINE/PubMed
BookMark	eNp9kc1P3DAQxa2KqnyUe6VKlSUuvWTr78THFbSwElWrsnCNvN4xGCV2sJMi_nu87JYDh55mNPq9mad5h2gvxAAIfaJkRinR35aL3zNGGJkxrSTT-h06oFrQihDB9kpPZF3VVOh9dJjzPSFUSKo-oH1eKKFreYAebjw8Vovw1yRvwojPAAY8T_bOj2DHKQF2MeGLqTcBz-3oY8B_wMbb4F_66-zDLV4-xupqTGB6_DO-zE1Y46s7MwBeQj_EZDp89hRM723-iN4702U43tUjdP3j-_L0orr8db44nV9WVlA-Vq54bay03IGwxKy5E7ZZ1ytuqLNlpAy3Wq04p1RLZQ1Qw51sauagsY1U_Ah93e4dUnyYII9t77OFrjMB4pRbxgVVjAmuC3ryBr2PUwrF3YbSlCkhWKG-7Khp1cO6HZLvTXpq_z2zAGQL2BRzTuBeEUraTV5tyavd5NXu8ioS9UZi_Wg2LxyT8d3_hJ-3Qg8Ar3carYikNX8GK5Kh8A
CODEN	IIPRE4
CitedBy_id	crossref_primary_10_1109_ACCESS_2022_3228825 crossref_primary_10_1109_TPAMI_2022_3183112 crossref_primary_10_1109_ACCESS_2024_3481631 crossref_primary_10_3233_JIFS_221468 crossref_primary_10_1109_JSEN_2022_3144157 crossref_primary_10_1016_j_compbiomed_2024_109399 crossref_primary_10_1007_s11042_022_13827_7 crossref_primary_10_1109_TCSVT_2021_3085959 crossref_primary_10_1016_j_asoc_2020_106859 crossref_primary_10_3389_fnbot_2024_1502071 crossref_primary_10_1177_02783649231207974 crossref_primary_10_1007_s11042_023_15443_5 crossref_primary_10_1145_3434746 crossref_primary_10_1109_JSEN_2023_3303912 crossref_primary_10_1117_1_JEI_30_3_033017 crossref_primary_10_1520_JTE20220704 crossref_primary_10_1016_j_jvcir_2022_103531 crossref_primary_10_1109_TIP_2023_3273459 crossref_primary_10_3389_fnbot_2024_1520983 crossref_primary_10_3390_diagnostics13030551 crossref_primary_10_1155_2022_2656001 crossref_primary_10_3389_fnins_2020_00409 crossref_primary_10_20965_jaciii_2024_p0552 crossref_primary_10_1016_j_imavis_2022_104403 crossref_primary_10_4218_etrij_2020_0101 crossref_primary_10_1109_ACCESS_2021_3099163 crossref_primary_10_1007_s11042_023_16001_9 crossref_primary_10_1109_TCSVT_2023_3255832 crossref_primary_10_1038_s41598_022_09293_8 crossref_primary_10_1109_JSEN_2022_3183502 crossref_primary_10_1007_s00521_023_09035_5 crossref_primary_10_3390_electronics10232901 crossref_primary_10_1007_s11042_022_14193_0 crossref_primary_10_3390_app12094165 crossref_primary_10_1007_s11760_023_02862_y crossref_primary_10_1007_s10462_024_10721_6 crossref_primary_10_1007_s13369_022_07236_z crossref_primary_10_1109_TCSVT_2022_3175959 crossref_primary_10_1016_j_imavis_2024_104919 crossref_primary_10_1007_s11042_023_16434_2 crossref_primary_10_1109_TCSVT_2021_3119956 crossref_primary_10_1109_TGRS_2022_3228959 crossref_primary_10_1016_j_bspc_2024_106870 crossref_primary_10_1109_JSEN_2024_3380321 crossref_primary_10_1007_s00530_024_01566_8 crossref_primary_10_1109_TCSVT_2022_3219864 crossref_primary_10_3390_s23020778 crossref_primary_10_1145_3679011 crossref_primary_10_1109_JSEN_2021_3122128 crossref_primary_10_1109_TCSVT_2022_3178430 crossref_primary_10_1016_j_ijepes_2022_108733 crossref_primary_10_1109_TCSVT_2023_3295432 crossref_primary_10_1007_s00138_024_01535_1 crossref_primary_10_1177_00368504211005480 crossref_primary_10_3233_AIC_220188 crossref_primary_10_1016_j_asoc_2025_112797 crossref_primary_10_1038_s41598_024_58190_9 crossref_primary_10_1109_TCSVT_2024_3417810 crossref_primary_10_1016_j_eswa_2022_118807 crossref_primary_10_1016_j_imavis_2023_104687 crossref_primary_10_1007_s11042_022_14214_y crossref_primary_10_1109_TCSVT_2021_3077512 crossref_primary_10_1007_s42452_021_04528_1 crossref_primary_10_1109_TBIOM_2022_3213545 crossref_primary_10_1109_TBIOM_2022_3216857 crossref_primary_10_1109_TCSVT_2022_3222305 crossref_primary_10_1109_TCSVT_2023_3249906 crossref_primary_10_1007_s11042_023_17345_y crossref_primary_10_1109_TCSVT_2021_3095290 crossref_primary_10_1109_TCSVT_2023_3236430 crossref_primary_10_1080_21681163_2021_2012829 crossref_primary_10_1080_21681163_2022_2111720 crossref_primary_10_1007_s00521_024_09630_0 crossref_primary_10_1007_s00371_023_03256_4 crossref_primary_10_1007_s10479_021_04164_3 crossref_primary_10_1016_j_image_2023_116940 crossref_primary_10_1016_j_bspc_2022_103772 crossref_primary_10_1109_ACCESS_2023_3298647 crossref_primary_10_1016_j_compbiomed_2021_105014 crossref_primary_10_1007_s10462_024_10934_9 crossref_primary_10_1109_TII_2023_3313640 crossref_primary_10_1109_TCSVT_2022_3194350 crossref_primary_10_1155_2023_9213689 crossref_primary_10_3233_JIFS_232842 crossref_primary_10_1016_j_neucom_2024_127882 crossref_primary_10_15622_ia_23_2_5 crossref_primary_10_1016_j_eswa_2023_122314 crossref_primary_10_1007_s00530_024_01312_0 crossref_primary_10_1007_s11042_020_09708_6 crossref_primary_10_1016_j_jvcir_2021_103161 crossref_primary_10_1038_s41467_023_39474_6 crossref_primary_10_3390_s22186841 crossref_primary_10_1007_s11042_022_12948_3 crossref_primary_10_3389_frsip_2024_1433388
Cites_doi	10.1109/CVPR.2015.7298860 10.1109/CVPR.2014.339 10.1109/ACCESS.2017.2753830 10.1109/ICCVW.2017.123 10.1109/TPAMI.2017.2691768 10.1186/s13640-017-0235-9 10.1109/CVPR.2018.00631 10.1109/TPAMI.2012.253 10.1109/CVPR.2016.331 10.1109/TCDS.2016.2577044 10.1109/CVPR.2016.167 10.1109/CVPRW.2017.207 10.1109/CVPR.2008.4587730 10.1007/s11042-017-4514-3 10.1109/TIP.2018.2818328 10.1016/j.patcog.2018.01.020 10.1109/ICCV.2013.394 10.1109/TIP.2003.819861 10.1016/j.patcog.2017.07.013 10.1109/CVPR.2013.347 10.1109/CVPR.2014.223 10.1109/34.910878 10.1109/5.726791 10.1007/s11263-019-01192-2 10.1007/s00138-010-0298-4 10.1016/j.ins.2018.12.050 10.1016/j.eswa.2018.03.047 10.1109/CVPR.2015.7299101 10.1109/CVPR.2015.7298691 10.1007/s10462-018-9651-1 10.1109/CVPR.2011.5995407 10.1023/B:STCO.0000035301.49549.88 10.1016/j.patcog.2017.06.037 10.1109/JSEN.2019.2903645 10.1016/j.cviu.2018.03.003 10.1109/TPAMI.2016.2533389 10.1109/DICTA.2017.8227505 10.1109/TIP.2017.2696786 10.1109/TPAMI.2012.59 10.1109/CVPR.2016.115 10.1109/CVPR.2015.7299176 10.1109/TCYB.2013.2265378 10.1109/CVPR.2014.333
ContentType	Journal Article
Copyright	Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2020
Copyright_xml	– notice: Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2020
DBID	97E RIA RIE AAYXX CITATION NPM 7SC 7SP 8FD JQ2 L7M L~C L~D 7X8
DOI	10.1109/TIP.2020.2965299
DatabaseName	IEEE Xplore (IEEE) 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: https://proxy.k.utb.cz/login?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://proxy.k.utb.cz/login?url=https://ieeexplore.ieee.org/ sourceTypes: Publisher
DeliveryMethod	fulltext_linktorsrc
Discipline	Applied Sciences Engineering
EISSN	1941-0042
EndPage	3844
ExternalDocumentID	31944975 10_1109_TIP_2020_2965299 8960517
Genre	orig-research Journal Article
GroupedDBID	--- -~X .DC 0R~ 29I 4.4 53G 5GY 5VS 6IK 97E AAJGR AARMG AASAJ AAWTH ABAZT ABFSI ABQJQ ABVLG ACGFO ACGFS ACIWK AENEX AETIX AGQYO AGSQL AHBIQ AI. AIBXA AKJIK AKQYR ALLEH ALMA_UNASSIGNED_HOLDINGS ASUFR ATWAV BEFXN BFFAM BGNUA BKEBE BPEOZ CS3 DU5 E.L EBS EJD F5P HZ~ H~9 ICLAB IFIPE IFJZH IPLJI JAVBF LAI M43 MS~ O9- OCL P2P RIA RIE RNS TAE TN5 VH1 AAYOK AAYXX CITATION RIG NPM Z5M 7SC 7SP 8FD JQ2 L7M L~C L~D 7X8
ID	FETCH-LOGICAL-c413t-f1458c5c3fe4c0ad3f4c8d7b3a1fce4c6a3c96b3311956cae1a3f5872fe8c8563
IEDL.DBID	RIE
ISSN	1057-7149 1941-0042
IngestDate	Fri Jul 11 15:26:17 EDT 2025 Mon Jun 30 10:25:08 EDT 2025 Wed Feb 19 02:09:36 EST 2025 Tue Jul 01 02:03:22 EDT 2025 Thu Apr 24 22:54:42 EDT 2025 Wed Aug 27 02:29:46 EDT 2025
IsPeerReviewed	true
IsScholarly	true
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-c413t-f1458c5c3fe4c0ad3f4c8d7b3a1fce4c6a3c96b3311956cae1a3f5872fe8c8563
Notes	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14 content type line 23
ORCID	0000-0002-1026-0047
PMID	31944975
PQID	2349126442
PQPubID	85429
PageCount	10
ParticipantIDs	proquest_journals_2349126442 crossref_primary_10_1109_TIP_2020_2965299 proquest_miscellaneous_2341622439 ieee_primary_8960517 pubmed_primary_31944975 crossref_citationtrail_10_1109_TIP_2020_2965299
ProviderPackageCode	CITATION AAYXX
PublicationCentury	2000
PublicationDate	2020-01-01
PublicationDateYYYYMMDD	2020-01-01
PublicationDate_xml	– month: 01 year: 2020 text: 2020-01-01 day: 01
PublicationDecade	2020
PublicationPlace	United States
PublicationPlace_xml	– name: United States – name: New York
PublicationTitle	IEEE transactions on image processing
PublicationTitleAbbrev	TIP
PublicationTitleAlternate	IEEE Trans Image Process
PublicationYear	2020
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	ref13 ref12 ref15 ref14 ref52 krizhevsky (ref37) 2012 ref10 ref17 ref19 ref18 liu (ref45) 2018; 1711 5941v4 ref50 ref46 jetley (ref28) 2014 song (ref48) 2017 ref47 ref42 ref41 ref44 ref43 ref8 han (ref2) 2013; 43 ref7 ref9 ref4 ref3 ref6 ref5 ref40 ref35 ref34 ref36 ref30 ref33 ref32 keçeli (ref51) 2018; 104 diba (ref11) 2017 ref1 ref39 liu (ref49) 2016 ref24 ref23 ref26 ref25 ref20 ref22 ref21 (ref38) 2016 ref27 ref29 lin (ref16) 2018 li (ref31) 2012
References_xml	– ident: ref33 doi: 10.1109/CVPR.2015.7298860 – ident: ref19 doi: 10.1109/CVPR.2014.339 – ident: ref15 doi: 10.1109/ACCESS.2017.2753830 – ident: ref36 doi: 10.1109/ICCVW.2017.123 – ident: ref44 doi: 10.1109/TPAMI.2017.2691768 – ident: ref9 doi: 10.1186/s13640-017-0235-9 – ident: ref22 doi: 10.1109/CVPR.2018.00631 – ident: ref10 doi: 10.1109/TPAMI.2012.253 – ident: ref43 doi: 10.1109/CVPR.2016.331 – year: 2012 ident: ref31 article-title: Cross-view activity recognition using Hankelets publication-title: Proc IEEE Conf Comput Vis Pattern Recognit – ident: ref12 doi: 10.1109/TCDS.2016.2577044 – ident: ref18 doi: 10.1109/CVPR.2016.167 – year: 2016 ident: ref49 article-title: Spatio-temporal LSTM with trust gates for 3D human action recognition publication-title: Proc Eur Conf Comput Vis (ECCV) – ident: ref50 doi: 10.1109/CVPRW.2017.207 – ident: ref40 doi: 10.1109/CVPR.2008.4587730 – ident: ref14 doi: 10.1007/s11042-017-4514-3 – year: 2018 ident: ref16 article-title: Action recognition with coarse-to-fine deep feature integration and asynchronous fusion publication-title: arXiv 1711 07430 – year: 2012 ident: ref37 article-title: ImageNet classification with deep convolutional neural networks publication-title: Proc Neural Inf Process Syst (NIPS) – ident: ref39 doi: 10.1109/TIP.2018.2818328 – ident: ref13 doi: 10.1016/j.patcog.2018.01.020 – ident: ref34 doi: 10.1109/ICCV.2013.394 – ident: ref41 doi: 10.1109/TIP.2003.819861 – ident: ref5 doi: 10.1016/j.patcog.2017.07.013 – ident: ref32 doi: 10.1109/CVPR.2013.347 – ident: ref8 doi: 10.1109/CVPR.2014.223 – start-page: 129 year: 2014 ident: ref28 article-title: 3D activity recognition using motion history and binary shape templates publication-title: Proc Asian Conf Comput Vis (ACCV) – year: 2016 ident: ref38 publication-title: Cmu motion capture database – ident: ref24 doi: 10.1109/34.910878 – ident: ref6 doi: 10.1109/5.726791 – ident: ref35 doi: 10.1007/s11263-019-01192-2 – year: 2017 ident: ref48 article-title: An end-to-end spatio-temporal attention model for human action recognition from skeleton data publication-title: Proc AAAI Conf Artif Intell – ident: ref29 doi: 10.1007/s00138-010-0298-4 – ident: ref23 doi: 10.1016/j.ins.2018.12.050 – volume: 104 start-page: 235 year: 2018 ident: ref51 article-title: Viewpoint projection based deep feature learning for single and dyadic action recognition publication-title: Expert Syst Appl doi: 10.1016/j.eswa.2018.03.047 – ident: ref27 doi: 10.1109/CVPR.2015.7299101 – ident: ref25 doi: 10.1109/CVPR.2015.7298691 – ident: ref3 doi: 10.1007/s10462-018-9651-1 – ident: ref47 doi: 10.1109/CVPR.2011.5995407 – ident: ref42 doi: 10.1023/B:STCO.0000035301.49549.88 – ident: ref7 doi: 10.1016/j.patcog.2017.06.037 – ident: ref4 doi: 10.1109/JSEN.2019.2903645 – ident: ref52 doi: 10.1016/j.cviu.2018.03.003 – ident: ref20 doi: 10.1109/TPAMI.2016.2533389 – ident: ref17 doi: 10.1109/DICTA.2017.8227505 – ident: ref30 doi: 10.1109/TIP.2017.2696786 – ident: ref1 doi: 10.1109/TPAMI.2012.59 – ident: ref21 doi: 10.1109/CVPR.2016.115 – ident: ref26 doi: 10.1109/CVPR.2015.7299176 – volume: 43 start-page: 1318 year: 2013 ident: ref2 article-title: Enhanced computer vision with microsoft Kinect sensor: A review publication-title: IEEE Trans Cybern doi: 10.1109/TCYB.2013.2265378 – ident: ref46 doi: 10.1109/CVPR.2014.333 – volume: 1711 5941v4 start-page: 1 year: 2018 ident: ref45 article-title: Skepxels: Spatio-temporal image representation of human skeleton joints for action recognition publication-title: CoRR – year: 2017 ident: ref11 article-title: Temporal 3D ConvNets: New architecture and transfer learning for video classification publication-title: arXiv 1711 08200
SSID	ssj0014516
Score	2.6269796
Snippet	Human action Recognition for unknown views, is a challenging task. We propose a deep view-invariant human action recognition framework, which is a novel...
SourceID	proquest pubmed crossref ieee
SourceType	Aggregation Database Index Database Enrichment Source Publisher
StartPage	3835
SubjectTerms	Algorithms Computer architecture Datasets Dynamics Human action recognition Human activity recognition Human motion human pose model Image recognition Invariants late fusion Shape Shape recognition spatial temporal dynamics Streaming media Trajectory Videos
Title	View-Invariant Deep Architecture for Human Action Recognition Using Two-Stream Motion and Shape Temporal Dynamics
URI	https://ieeexplore.ieee.org/document/8960517 https://www.ncbi.nlm.nih.gov/pubmed/31944975 https://www.proquest.com/docview/2349126442 https://www.proquest.com/docview/2341622439
Volume	29
hasFullText	1
inHoldings	1
isFullTextHit
isPrint
link	http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwjV3db9MwED9te0DjgcHGR2FDRuIFCbdJHDvxY8U2bUhFCHVob5HtXLSJLe3Wlmn76zk7H6oQIN6sxLGd_M7x73znO4D3nvVaTATHzEhO-obhVlUlz6xSqY2rMgtnqyZf1MlZ-vlcnm_Ax_4sDCIG5zMc-mKw5Zczt_JbZaOc6LaMs03YJMWtOavVWwx8wtlg2ZQZz4j2dybJSI-mp19JEUyiYaKVpN_vNjwiwUtT7Z0L11ajkF7l70wzrDjHOzDpxto4mvwYrpZ26B5-C-P4vy_zFJ601JONG1l5BhtY78JOS0NZO8kXu_B4LUbhHtx8v8Q7flr_JKWaUGCHiHM2XjM_MKK9LNgC2DgckmDfOqckKgeXBDa9m3Fv_zbXbBLSBjFTU5cXZo5s2gTHumKH97W5vnSL53B2fDT9dMLbPA3c0RK45BV9_NxJJypMXWRKUaUuLzMrTFw5uqSMcFpZIXx4OeUMxkZUMs-SCnOXSyVewFY9q_EVMEKJCJzVLnFRalKpldfQfFQzgdoYOYBRh1fh2iDmPpfGVRGUmUgXBHbhwS5asAfwoX9i3gTw-EfdPY9TX6-FaAD7nUgU7QxfFIlIdezZZDKAd_1tmpve4GJqnK1CnVgRRxLU8stGlPq2Owl8_ec-38C2H1mz2bMPW8vbFR4Q_Vnat0HufwHkB_4F
linkProvider	IEEE
linkToHtml	http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwjV3db9MwED-NIcF4YLANKAwwEi9IuE3ij8SPFWNqYZ0QytDeIttxxMSWFtoywV_P2flQhQDxZiWO7eTu4t_5vgBeetRrXMKoS7WgqG9oamRV0tRIyU1clWmIrZqdyskZf3cuzrfgdR8L45wLzmdu6JvBll_O7doflY0yhNsiTm_ATdz3RdxEa_U2A19yNtg2RUpTBP6dUTJSo3z6AVXBJBomSgr8Ae_ALWQ9zpV3L9zYj0KBlb9jzbDnHO_CrFtt42ryZbhemaH9-Vsix_99nXtwtwWfZNxwy33YcvUe7LZAlLRivtyDOxtZCvfh66cLd02n9XdUq5EO5Mi5BRlvGCAIAl8SrAFkHMIkyMfOLQnbwSmB5Ndz6i3g-orMQuEgomuc8rNeOJI36bEuydGPWl9d2OUBnB2_zd9MaFupgVrcBFe0wo-fWWFZ5biNdMkqbrMyNUzHlcVLUjOrpGHMJ5iTVrtYs0pkaVK5zGZCsgewXc9r9wgIUgkhnFE2sRHXXCjpdTSf14w5pbUYwKijV2HbNOa-msZlEdSZSBVI7MITu2iJPYBX_ROLJoXHP_ruezr1_VoSDeCwY4milfFlkTCuYo8nkwG86G-jdHqTi67dfB36xBJREsORHzas1I_dceDjP8_5HG5P8tlJcTI9ff8Edvwqm6OfQ9hefVu7pwiGVuZZkIFfSNsBXQ
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=View-Invariant+Deep+Architecture+for+Human+Action+Recognition+Using+Two-Stream+Motion+and+Shape+Temporal+Dynamics&rft.jtitle=IEEE+transactions+on+image+processing&rft.au=Dhiman%2C+Chhavi&rft.au=Vishwakarma%2C+Dinesh+Kumar&rft.date=2020-01-01&rft.issn=1057-7149&rft.eissn=1941-0042&rft.volume=29&rft.spage=3835&rft.epage=3844&rft_id=info:doi/10.1109%2FTIP.2020.2965299&rft.externalDBID=n%2Fa&rft.externalDocID=10_1109_TIP_2020_2965299
thumbnail_l	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1057-7149&client=summon
thumbnail_m	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1057-7149&client=summon
thumbnail_s	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1057-7149&client=summon