Conference item icon

Conference item

Sim2real transfer learning for 3D human pose estimation: motion to the rescue

Abstract:

Synthetic visual data can provide practicically infinite diversity and rich labels, while avoiding ethical issues with privacy and bias. However, for many tasks, current models trained on synthetic data generalize poorly to real data. The task of 3D human pose estimation is a particularly interesting example of this sim2real problem, because learning-based approaches perform reasonably well given real training data, yet labeled 3D poses are extremely difficult to obtain in the wild, limiting ...

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

Actions


Access Document


Files:
Publication website:
http://www.proceedings.com/53719.html

Authors


More by this author
Department:
ENGINEERING SCIENCE
Sub department:
Engineering Science
Oxford college:
Brasenose College
Role:
Author
ORCID:
0000-0002-8945-8573
Publisher:
Curran Associates Publisher's website
Volume:
32
Pages:
12905-12917
Series:
Advances in Neural Information Processing Systems
Series number:
32
Host title:
Advances in Neural Information Processing Systems 32: 32nd Conference on Neural Information Processing Systems (NeurIPS 2019)
Publication date:
2020-06-01
Acceptance date:
2019-09-04
Event title:
Thirty-third Conference on Neural Information Processing Systems (NeurIPS | 2019)
Event location:
Vancouver, Canada
Event website:
https://nips.cc/Conferences/2019
Event start date:
2019-12-08T00:00:00Z
Event end date:
2019-12-14T00:00:00Z
ISSN:
1049-5258
ISBN:
9781713807933
Language:
English
Keywords:
Pubs id:
1118223
Local pid:
pubs:1118223
Deposit date:
2020-07-20

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