TY - JOUR
T1 - Convergence of Wachspress coordinates: from polygons to curved domains
AU - Kosinka, Jiří
AU - Barton, Michael
N1 - KAUST Repository Item: Exported on 2020-10-01
PY - 2014/8/8
Y1 - 2014/8/8
N2 - Given a smooth, strictly convex planar domain, we investigate point-wise convergence of the sequence of Wachspress coordinates defined over finer and finer inscribed polygonal approximations of the domain. Based on a relation between the discrete Wachspress case and the limit smooth case given by the Wachspress kernel defined by Warren et al., we show that the corresponding sequences of Wachspress interpolants and mappings converge as 𝓞(h2) for a sampling step size h of the boundary curve of the domain as h → 0. Several examples are shown to numerically validate the results and to visualise the behaviour of discrete interpolants and mappings as they converge to their smooth counterparts. Empirically, the same convergence order is observed also for mean value coordinates. Moreover, our numerical tests suggest that the convergence of interpolants and mappings is uniform both in the Wachspress and mean value cases. © 2014 Springer Science+Business Media New York.
AB - Given a smooth, strictly convex planar domain, we investigate point-wise convergence of the sequence of Wachspress coordinates defined over finer and finer inscribed polygonal approximations of the domain. Based on a relation between the discrete Wachspress case and the limit smooth case given by the Wachspress kernel defined by Warren et al., we show that the corresponding sequences of Wachspress interpolants and mappings converge as 𝓞(h2) for a sampling step size h of the boundary curve of the domain as h → 0. Several examples are shown to numerically validate the results and to visualise the behaviour of discrete interpolants and mappings as they converge to their smooth counterparts. Empirically, the same convergence order is observed also for mean value coordinates. Moreover, our numerical tests suggest that the convergence of interpolants and mappings is uniform both in the Wachspress and mean value cases. © 2014 Springer Science+Business Media New York.
UR - http://hdl.handle.net/10754/563698
UR - http://link.springer.com/10.1007/s10444-014-9370-3
UR - http://www.scopus.com/inward/record.url?scp=84931565761&partnerID=8YFLogxK
U2 - 10.1007/s10444-014-9370-3
DO - 10.1007/s10444-014-9370-3
M3 - Article
SN - 1019-7168
VL - 41
SP - 489
EP - 505
JO - Advances in Computational Mathematics
JF - Advances in Computational Mathematics
IS - 3
ER -