PoTion: Pose MoTion Representation for Action Recognition

Most state-of-the-art methods for action recognition rely on a two-stream architecture that processes appearance and motion independently. In this paper, we claim that considering them jointly offers rich information for action recognition. We introduce a novel representation that gracefully encodes...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition s. 7024 - 7033
Hlavní autori: Choutas, Vasileios, Weinzaepfel, Philippe, Revaud, Jerome, Schmid, Cordelia
Médium: Konferenčný príspevok..
Jazyk:English
Vydavateľské údaje: IEEE 01.06.2018
Predmet:
ISSN:1063-6919
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Abstract Most state-of-the-art methods for action recognition rely on a two-stream architecture that processes appearance and motion independently. In this paper, we claim that considering them jointly offers rich information for action recognition. We introduce a novel representation that gracefully encodes the movement of some semantic keypoints. We use the human joints as these keypoints and term our Pose moTion representation PoTion. Specifically, we first run a state-of-the-art human pose estimator [4] and extract heatmaps for the human joints in each frame. We obtain our PoTion representation by temporally aggregating these probability maps. This is achieved by 'colorizing' each of them depending on the relative time of the frames in the video clip and summing them. This fixed-size representation for an entire video clip is suitable to classify actions using a shallow convolutional neural network. Our experimental evaluation shows that PoTion outperforms other state-of-the-art pose representations [6, 48]. Furthermore, it is complementary to standard appearance and motion streams. When combining PoTion with the recent two-stream I3D approach [5], we obtain state-of-the-art performance on the JHMDB, HMDB and UCF101 datasets.
AbstractList Most state-of-the-art methods for action recognition rely on a two-stream architecture that processes appearance and motion independently. In this paper, we claim that considering them jointly offers rich information for action recognition. We introduce a novel representation that gracefully encodes the movement of some semantic keypoints. We use the human joints as these keypoints and term our Pose moTion representation PoTion. Specifically, we first run a state-of-the-art human pose estimator [4] and extract heatmaps for the human joints in each frame. We obtain our PoTion representation by temporally aggregating these probability maps. This is achieved by 'colorizing' each of them depending on the relative time of the frames in the video clip and summing them. This fixed-size representation for an entire video clip is suitable to classify actions using a shallow convolutional neural network. Our experimental evaluation shows that PoTion outperforms other state-of-the-art pose representations [6, 48]. Furthermore, it is complementary to standard appearance and motion streams. When combining PoTion with the recent two-stream I3D approach [5], we obtain state-of-the-art performance on the JHMDB, HMDB and UCF101 datasets.
Author Weinzaepfel, Philippe
Revaud, Jerome
Choutas, Vasileios
Schmid, Cordelia
Author_xml – sequence: 1
  givenname: Vasileios
  surname: Choutas
  fullname: Choutas, Vasileios
– sequence: 2
  givenname: Philippe
  surname: Weinzaepfel
  fullname: Weinzaepfel, Philippe
– sequence: 3
  givenname: Jerome
  surname: Revaud
  fullname: Revaud, Jerome
– sequence: 4
  givenname: Cordelia
  surname: Schmid
  fullname: Schmid, Cordelia
BookMark eNotjLFOwzAURQ0CibZkZmDJDyR9tuNnm62KKCAVEVWFtbKdFwiCuIqz8PeolOncoyudObsY4kCM3XAoOQe7rN-abSmAmxJAy-qMZVYbrqRBrATYczbjgLJAy-0Vy1L6BACBRppKzZht4q6Pw13exET585_kWzqMlGiY3HTULo75KkynJ8T3oT_ua3bZua9E2T8X7HV9v6sfi83Lw1O92hQfQsFUoCTkFXaBKtFB8AEVetcCoOPgCEwgrzsSLbbGIgrrtRdeEkntKIhWLtjtqdsT0f4w9t9u_NkbpY2RQv4CpBNJxg
CODEN IEEPAD
ContentType Conference Proceeding
DBID 6IE
6IH
CBEJK
RIE
RIO
DOI 10.1109/CVPR.2018.00734
DatabaseName IEEE Electronic Library (IEL) Conference Proceedings
IEEE Proceedings Order Plan (POP) 1998-present by volume
IEEE Xplore All Conference Proceedings
IEEE Electronic Library (IEL)
IEEE Proceedings Order Plans (POP) 1998-present
DatabaseTitleList
Database_xml – sequence: 1
  dbid: RIE
  name: IEEE Electronic Library (IEL)
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
Discipline Applied Sciences
EISBN 9781538664209
1538664208
EISSN 1063-6919
EndPage 7033
ExternalDocumentID 8578832
Genre orig-research
GroupedDBID 6IE
6IH
6IL
6IN
AAWTH
ABLEC
ADZIZ
ALMA_UNASSIGNED_HOLDINGS
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CBEJK
CHZPO
IEGSK
IJVOP
OCL
RIE
RIL
RIO
ID FETCH-LOGICAL-h250t-63e6146fce42f0cbc656bad006a10ae08ceb7fe2d6d896629b7b2b3ee37aec2d3
IEDL.DBID RIE
ISICitedReferencesCount 232
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000457843607019&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
IngestDate Wed Aug 27 02:52:16 EDT 2025
IsDoiOpenAccess false
IsOpenAccess true
IsPeerReviewed false
IsScholarly true
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-h250t-63e6146fce42f0cbc656bad006a10ae08ceb7fe2d6d896629b7b2b3ee37aec2d3
OpenAccessLink https://inria.hal.science/hal-01764222
PageCount 10
ParticipantIDs ieee_primary_8578832
PublicationCentury 2000
PublicationDate 2018-Jun
PublicationDateYYYYMMDD 2018-06-01
PublicationDate_xml – month: 06
  year: 2018
  text: 2018-Jun
PublicationDecade 2010
PublicationTitle 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition
PublicationTitleAbbrev CVPR
PublicationYear 2018
Publisher IEEE
Publisher_xml – name: IEEE
SSID ssj0002683845
ssj0003211698
Score 2.570884
Snippet Most state-of-the-art methods for action recognition rely on a two-stream architecture that processes appearance and motion independently. In this paper, we...
SourceID ieee
SourceType Publisher
StartPage 7024
SubjectTerms Computer architecture
Heating systems
Image color analysis
Joints
Optical imaging
Streaming media
Title PoTion: Pose MoTion Representation for Action Recognition
URI https://ieeexplore.ieee.org/document/8578832
WOSCitedRecordID wos000457843607019&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV09T8MwED21iIGpQIv4lgdGTNM4sR02VFGxUEVVQd0q2zkLlgT1g9_POQmFgYUtjhTLes7pnu33fAA3xLkp7xQFF6kzPEl8wm1qU66twJC_rDZNsQk1nerFIss7cLvzwiBiLT7Du_BYn-UXlduGrbKhpn7pD-xCVynZeLV2-ymx1EK3J2ShLWhlIzPd3uYzirLh-DWfBS1XEE-qUCj5VzmVOptMev8bxyEMfmx5LN8lnCPoYHkMvZZHsjZK133I8mpOcN-zvFoje64bbFZLXlunUcmIq7KH2tPAZt8aoqocwMvkcT5-4m2JBP5G3GXDpUDKr9I7TGIfOeuInllTUCiZUWQw0g6t8hgXstC0sIkzq2xM04BCGXRxIU5gr6xKPAXmkBD03ikUFNipMPS5M0lmvbSZ9PYM-gGJ5UdzC8ayBeH879cXcBCgbkRVl7C3WW3xCvbd5-Z9vbqup-4LL5-aIg
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV09T8MwED2VggRTgRbxjQdGQoOdD4cNVVRFtFVUFdStsp2zYElQP_j9nJNQGFjY4kixrOec7tl-zwdwTZyb8k6WeSI0ygsCG3g61KEntUCXv7RUVbGJeDyWs1mSNuBm44VBxFJ8hrfusTzLzwqzdltlXUn90h-4BdthEHC_cmttdlR4JIWsz8hcW9DaJkpkfZ_PnZ90e6_pxKm5nHwydqWSfxVUKfNJv_W_kexD58eYx9JNyjmABuaH0KqZJKvjdNmGJC2mBPg9S4slslHZYJNS9Fp7jXJGbJU9lK4GNvlWERV5B176j9PewKuLJHhvxF5WXiSQMmxkDQbc-kYbImhaZRRM6s5X6EuDOrbIsyiTtLThiY41p4lAESs0PBNH0MyLHI-BGSQErTUxCgrtUCj63Kgg0TbSSWT1CbQdEvOP6h6MeQ3C6d-vr2B3MB0N58On8fMZ7DnYK4nVOTRXizVewI75XL0vF5flNH4BZFqdaQ
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%3Abook&rft.genre=proceeding&rft.title=2018+IEEE%2FCVF+Conference+on+Computer+Vision+and+Pattern+Recognition&rft.atitle=PoTion%3A+Pose+MoTion+Representation+for+Action+Recognition&rft.au=Choutas%2C+Vasileios&rft.au=Weinzaepfel%2C+Philippe&rft.au=Revaud%2C+Jerome&rft.au=Schmid%2C+Cordelia&rft.date=2018-06-01&rft.pub=IEEE&rft.eissn=1063-6919&rft.spage=7024&rft.epage=7033&rft_id=info:doi/10.1109%2FCVPR.2018.00734&rft.externalDocID=8578832