A belief-theoretical approach to example-based pose estimation
- Submitting institution
-
Oxford Brookes University
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 185738365
- Type
- D - Journal article
- DOI
-
10.1109/TFUZZ.2017.2686803
- Title of journal
- IEEE Transactions on Fuzzy Systems
- Article number
- -
- First page
- 598
- Volume
- 26
- Issue
- 2
- ISSN
- 1063-6706
- Open access status
- Compliant
- Month of publication
- March
- Year of publication
- 2017
- 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
-
1
- Research group(s)
-
-
- Citation count
- 0
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- This paper is a rare example of the application of uncertainty theory, and belief function theory in particular, to regression problems, in particular the task of estimating the pose of an articulated objects from an image sequence. By virtue of its originality it was published on a very high impact journal, IEEE Fuzzy Systems (in 2018 the highest-impact-factor journal in the whole of computer science). Its significance lies in showing, arguably for the first time, that a regression framework modelling higher-order uncertainty can outperform mainstream machine learning methods such as Gaussian processes and RVMs.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -