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:
-
-
(Accepted manuscript, 5.0MB)
-
- Publication website:
- http://www.proceedings.com/53719.html
Authors
Bibliographic Details
- Publisher:
- Curran Associates Publisher's website
- Host title:
- Advances in Neural Information Processing Systems 32: 32nd Conference on Neural Information Processing Systems (NeurIPS 2019)
- Series:
- Advances in Neural Information Processing Systems
- Series number:
- 32
- Volume:
- 32
- Pages:
- 12905-12917
- 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:
- 2019-12-08
- Event end date:
- 2019-12-14
- ISSN:
-
1049-5258
- ISBN:
- 9781713807933
Item Description
- Language:
- English
- Keywords:
- Pubs id:
-
1118223
- Local pid:
- pubs:1118223
- Deposit date:
- 2020-07-20
Terms of use
- Copyright holder:
- Doersch and Zisserman
- Copyright date:
- 2019
- Notes:
- This is the accepted manuscript version of the paper. The final version is available online from NIPS at https://papers.nips.cc/paper/9454-sim2real-transfer-learning-for-3d-human-pose-estimation-motion-to-the-rescue
Metrics
If you are the owner of this record, you can report an update to it here: Report update to this record