A Sampling Approach to Generating Closely Interacting 3D Pose-pairs from 2D Annotations
- Submitting institution
-
University of Northumbria at Newcastle
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 22062494
- Type
- D - Journal article
- DOI
-
10.1109/TVCG.2018.2832097
- Title of journal
- IEEE Transactions on Visualization and Computer Graphics
- Article number
- -
- First page
- 2217
- Volume
- 25
- Issue
- 6
- ISSN
- 1077-2626
- Open access status
- Compliant
- Month of publication
- May
- Year of publication
- 2018
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
6
- Research group(s)
-
D - Computer Vision and Natural Computing (CVNC)
- Citation count
- 1
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- To enable people to access to a huge amount of data synthesized from a few manually annotated posture-pairs by our new MCMC sampling approach, we published a huge amount (over 5GBytes) of close interactions data online as a community resource (URL: https://www.dropbox.com/s/nvxwi84da9vej6d/pairint_annotation.zip?dl=1). The key advance in generating a large volume of close human interaction motion data led to the potential applications in modelling interaction using deep learning frameworks. Dr Edmond Ho was invited to give a research talk on this topic at the School of Computing, University of Portsmouth (Dr Zhaojie Ju zhaojie.ju@port.ac.uk, URL: https://www.portaisociety.com/researchseminars/) in October 2018.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -