Highly articulated kinematic structure estimation combining motion and skeleton information
- Submitting institution
-
The University of Birmingham
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 48683051
- Type
- D - Journal article
- DOI
-
10.1109/TPAMI.2017.2748579
- Title of journal
- IEEE Transactions on Pattern Analysis and Machine Intelligence
- Article number
- -
- First page
- 2165
- Volume
- 40
- Issue
- 9
- ISSN
- 0162-8828
- Open access status
- Technical exception
- Month of publication
- September
- 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
- 2
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- This paper addresses a novel problem of estimating arbitrary objects' kinematic structure from videos without any prior knowledge of the objects. It presents a new methodology that is the basis for robots to learn and manipulate objects by recognising motion and structural information of objects. The proposed method was implemented in the humanoid-robot(iCub) and was successfully used by the robot to recognise its hand structure (EU FP7 project WYSIWYD under Grant 612139). The paper was awarded the University of Birmingham EPS College Paper of the Month award.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -