Provably good moving least squares Export

@inproceedings{citeulike:266063, address = {Philadelphia, PA, USA}, author = {Kolluri, Ravikrishna}, booktitle = {SODA '05: Proceedings of the sixteenth annual ACM-SIAM symposium on Discrete algorithms}, citeulike-article-id = {266063}, citeulike-linkout-0 = {http://portal.acm.org/citation.cfm?id=1070432.1070578}, comment = {This paper describes a variation of Moving Least Square. The require global \$\epsilon\$ sampling (lower bounded by the smallest local feature size). They show that, for a point \$x\$, the sign of the signed distance is correct, if \$x\$ is not very close to the surface. This is quite interesting. Also, because they do not allow point arbitrary close to the surface, the implicit function can be reduced to use LOCAL neighbors instead of the entire point set. Which is helpful in time-complexity. ---=note-separator=--- Though I am not sure whether the global \$\epsilon\$ sampling can be replaced by normal \$\epsilon\$ sampling, or \$(\epsilon,\delta)\$ sampling.}, isbn = {0898715857}, keywords = {graphics, mesh, point, reconstruction, surface}, pages = {1008--1017}, posted-at = {2005-07-27 08:08:11}, priority = {0}, publisher = {Society for Industrial and Applied Mathematics}, title = {Provably good moving least squares}, url = {http://portal.acm.org/citation.cfm?id=1070432.1070578}, year = {2005} }

TY - CONF ID - citeulike:266063 L3 - citeulike-article-id:266063 SN - 0898715857 AD - Philadelphia, PA, USA EP - 1017 TI - Provably good moving least squares PB - Society for Industrial and Applied Mathematics T2 - SODA '05: Proceedings of the sixteenth annual ACM-SIAM symposium on Discrete algorithms SP - 1008 KW - graphics KW - mesh KW - point KW - reconstruction KW - surface N1 - This paper describes a variation of Moving Least Square. The require global $\epsilon$ sampling (lower bounded by the smallest local feature size). They show that, for a point $x$, the sign of the signed distance is correct, if $x$ is not very close to the surface. This is quite interesting. Also, because they do not allow point arbitrary close to the surface, the implicit function can be reduced to use LOCAL neighbors instead of the entire point set. Which is helpful in time-complexity. N1 - Though I am not sure whether the global $\epsilon$ sampling can be replaced by normal $\epsilon$ sampling, or $(\epsilon,\delta)$ sampling. AU - Kolluri, Ravikrishna PY - 2005/// UR - http://portal.acm.org/citation.cfm?id=1070432.1070578 ER -