Conference item icon

Conference item

Quo Vadis, action recognition? A new model and the kinetics dataset

Abstract:

The paucity of videos in current action classification datasets (UCF-101 and HMDB-51) has made it difficult to identify good video architectures, as most methods obtain similar performance on existing small-scale benchmarks. This paper re-evaluates state-of-the-art architectures in light of the new Kinetics Human Action Video dataset. Kinetics has two orders of magnitude more data, with 400 human action classes and over 400 clips per class, and is collected from realistic, challenging YouTube...

Expand abstract
Publication status:
Published
Peer review status:
Peer reviewed

Actions


Access Document


Files:
Publisher copy:
10.1109/cvpr.2017.502

Authors


More by this author
Institution:
University of Oxford
Division:
MPLS
Department:
Engineering Science
Oxford college:
Brasenose College
Role:
Author
ORCID:
0000-0002-8945-8573
Publisher:
IEEE Publisher's website
Journal:
2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Journal website
Host title:
2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
Publication date:
2017-11-09
Acceptance date:
2017-03-03
Event location:
Honolulu, Hawaii
DOI:
ISSN:
1063-6919
Source identifiers:
820333
Pubs id:
pubs:820333
UUID:
uuid:c8881c3e-c91d-444c-b142-75f9ff6377af
Local pid:
pubs:820333
Deposit date:
2019-01-29

Terms of use


Views and Downloads






If you are the owner of this record, you can report an update to it here: Report update to this record

TO TOP