Evaluating the covariance matrix constraints for data-driven statistical human motion reconstruction
- Submitting institution
-
University of Sussex
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 126552_47155
- Type
- E - Conference contribution
- DOI
-
10.1145/2643188.2643199
- Title of conference / published proceedings
- SCCG '14: Proceedings of the 30th Spring Conference on Computer Graphics
- First page
- 99
- Volume
- -
- Issue
- -
- ISSN
- -
- Open access status
- -
- Month of publication
- May
- Year of publication
- 2014
- URL
-
http://doi.acm.org/10.1145/2643188.2643199
- 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
- Yes
- Number of additional authors
-
2
- Research group(s)
-
-
- Citation count
- -
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- This research develops a novel technique for efficient data driven animation from motion capture data and is the first significant comparative study of how four specific covariance matrix constraints influence the motion reconstruction process. This work is significant in its demonstration of realistic character motion simulation, showing that FA and PCA reconstruct higher quality motion sequences with reduced measurement error compared to other constrained approaches and that this can be applied to a restricted set of markers. In particular the restricted markers contribution has become a key benchmark for work in this area. Field-weighted citation impact 1.48 (Scopus).
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -