Non-rigid registration under anisotropic deformations
- Submitting institution
-
Swansea University / Prifysgol Abertawe
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 49861
- Type
- D - Journal article
- DOI
-
10.1016/j.cagd.2019.04.014
- Title of journal
- Computer Aided Geometric Design
- Article number
- -
- First page
- 142
- Volume
- 71
- Issue
- -
- ISSN
- 0167-8396
- Open access status
- Compliant
- Month of publication
- May
- Year of publication
- 2019
- 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
-
-
- Research group(s)
-
-
- Citation count
- 3
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- Real-life 3D scans are challenging to register. This study shows that real-life scans with large-scale pose differences often contain anisotropic non-isometric deformation. This paper introduces the first* registration technique that explicitly models anisotropic, non-isometric and large-scale deformation all together. The modelling of anisotropic deformation helps to register scans especially in joint areas. It enables accurate registration of real-life, arbitrary* 3D scans whilst being fast*. It out-performed 5 previous best techniques on 5 public (synthesis and real) datasets, and is robust against holes and cuts. It underpinned a new SHREC’20 dataset (correspondence with non-isometric deformation) enabling subsequent anisotropic/non-isometric research.(*independent survey doi:10.1007/s00371-019-01760-0)
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -