A novel symmetric skew-Hamiltonian isotropic Lanczos algorithm for spectral conformal parameterizations (Q487705): Difference between revisions

Within the context of geometry processing, the authors deal with spectral conformal parametrization (SCP), which is one of the methods used to compute a quality conformal parametrization based on the spectral techniques. This paper provides three main contributions: 1) nonequivalence deflation, i.e., a deflation technique which transforms the zero eigenvalues of a general eigenvalue problem (GEP) into the infinite ones while preserving all other eigenvalues; 2) null-space free compression, which reduces a GEP to a small-scale standard eigenvalue problem (SEP), based on the low-rank property; 3) \(\mathbb{S}\)HILA algorithm, which is a new symmetric skew-Hamiltonian isotropic Lanczos algorithm to solve the reduced problem. The result is a new, efficient, accurate and robust eigensolver for the SCP. Some numerical examples enlighten these features.