Publication
Human action recognition in videos with articulated pose information by deep networks
dc.contributor.author | Farrajota, Miguel | |
dc.contributor.author | Rodrigues, João | |
dc.contributor.author | du Buf, J. M. H. | |
dc.date.accessioned | 2020-07-24T10:51:09Z | |
dc.date.available | 2020-07-24T10:51:09Z | |
dc.date.issued | 2019-11 | |
dc.description.abstract | Action recognition is of great importance in understanding human motion from video. It is an important topic in computer vision due to its many applications such as video surveillance, human-machine interaction and video retrieval. One key problem is to automatically recognize low-level actions and high-level activities of interest. This paper proposes a way to cope with low-level actions by combining information of human body joints to aid action recognition. This is achieved by using high-level features computed by a convolutional neural network which was pre-trained on Imagenet, with articulated body joints as low-level features. These features are then used to feed a Long Short-Term Memory network to learn the temporal dependencies of an action. For pose prediction, we focus on articulated relations between body joints. We employ a series of residual auto-encoders to produce multiple predictions which are then combined to provide a likelihood map of body joints. In the network topology, features are processed across all scales which capture the various spatial relationships associated with the body. Repeated bottom-up and top-down processing with intermediate supervision of each auto-encoder network is applied. We demonstrate state-of-the-art results on the popular FLIC, LSP and UCF Sports datasets. | |
dc.description.sponsorship | FCT Project LARSyS [UID/EEA/50009/2013] | |
dc.description.sponsorship | FCT Ph.D. GrantPortuguese Foundation for Science and Technology [SFRH/BD/79812/2011] | |
dc.description.version | info:eu-repo/semantics/publishedVersion | |
dc.identifier.doi | 10.1007/s10044-018-0727-y | |
dc.identifier.issn | 1433-7541 | |
dc.identifier.uri | http://hdl.handle.net/10400.1/14203 | |
dc.language.iso | eng | |
dc.peerreviewed | yes | |
dc.publisher | Springer | |
dc.relation | NEURAL CORRELATES OF MOTION AND STEREO VISION IN HUMAN POSE AND GAIT DETECTION | |
dc.subject | Human action | |
dc.subject | Human pose | |
dc.subject | ConvNet | |
dc.subject | Neural networks | |
dc.subject | Auto-encoders | |
dc.subject | LSTM | |
dc.title | Human action recognition in videos with articulated pose information by deep networks | |
dc.type | journal article | |
dspace.entity.type | Publication | |
oaire.awardTitle | NEURAL CORRELATES OF MOTION AND STEREO VISION IN HUMAN POSE AND GAIT DETECTION | |
oaire.awardURI | info:eu-repo/grantAgreement/FCT/5876/UID%2FEEA%2F50009%2F2013/PT | |
oaire.awardURI | info:eu-repo/grantAgreement/FCT//SFRH%2FBD%2F79812%2F2011/PT | |
oaire.citation.endPage | 1318 | |
oaire.citation.issue | 4 | |
oaire.citation.startPage | 1307 | |
oaire.citation.title | Pattern Analysis and Applications | |
oaire.citation.volume | 22 | |
oaire.fundingStream | 5876 | |
person.familyName | Farrajota | |
person.familyName | Rodrigues | |
person.familyName | du Buf | |
person.givenName | Miguel | |
person.givenName | Joao | |
person.givenName | Hans | |
person.identifier.ciencia-id | 8A19-98F7-9914 | |
person.identifier.orcid | 0000-0001-7970-4649 | |
person.identifier.orcid | 0000-0002-3562-6025 | |
person.identifier.orcid | 0000-0002-4345-1237 | |
person.identifier.rid | M-5125-2013 | |
person.identifier.scopus-author-id | 55807461600 | |
person.identifier.scopus-author-id | 6604075916 | |
project.funder.identifier | http://doi.org/10.13039/501100001871 | |
project.funder.identifier | http://doi.org/10.13039/501100001871 | |
project.funder.name | Fundação para a Ciência e a Tecnologia | |
project.funder.name | Fundação para a Ciência e a Tecnologia | |
rcaap.rights | restrictedAccess | |
rcaap.type | article | |
relation.isAuthorOfPublication | ca40db25-2109-4390-ad94-c5f28803c7e8 | |
relation.isAuthorOfPublication | 683ba85b-459c-4789-a4ff-a4e2a904b295 | |
relation.isAuthorOfPublication | cfad5636-2c77-4db0-a3a0-d7eb97ce6bee | |
relation.isAuthorOfPublication.latestForDiscovery | 683ba85b-459c-4789-a4ff-a4e2a904b295 | |
relation.isProjectOfPublication | 19a727e2-d775-407f-ab10-7d5f19577e08 | |
relation.isProjectOfPublication | d150d21b-0a3e-4403-92cb-65f221ecb683 | |
relation.isProjectOfPublication.latestForDiscovery | 19a727e2-d775-407f-ab10-7d5f19577e08 |
Files
Original bundle
1 - 1 of 1
No Thumbnail Available
- Name:
- Farrajota2019_Article_HumanActionRecognitionInVideos.pdf
- Size:
- 2.65 MB
- Format:
- Adobe Portable Document Format