Bein, Matthias ; Fellner, Dieter W. ; Stork, André (2011)
Genetic B-Spline Approximation on Combined B-Reps.
In: The Visual Computer, 27 (6-8)
doi: 10.1007/s00371-011-0592-9
Artikel, Bibliographie
Kurzbeschreibung (Abstract)
We present a genetic algorithm for approximating densely sampled curves with uniform cubic B-Splines suitable for Combined B-reps. A feature of this representation is altering the continuity property of the B-Spline at any knot, allowing combining freeform curves and polygonal parts within one representation. Naturally there is a trade-off between different approximation properties like accuracy and the number of control points needed. Our algorithm creates very accurate B-Splines with few control points, as shown in Fig. 1. Since the approximation problem is highly nonlinear, we approach it with genetic methods, leading to better results compared to classical gradient based methods. Parallelization and adapted evolution strategies are used to create results very fast.
| Typ des Eintrags: | Artikel |
|---|---|
| Erschienen: | 2011 |
| Autor(en): | Bein, Matthias ; Fellner, Dieter W. ; Stork, André |
| Art des Eintrags: | Bibliographie |
| Titel: | Genetic B-Spline Approximation on Combined B-Reps |
| Sprache: | Englisch |
| Publikationsjahr: | 2011 |
| Titel der Zeitschrift, Zeitung oder Schriftenreihe: | The Visual Computer |
| Jahrgang/Volume einer Zeitschrift: | 27 |
| (Heft-)Nummer: | 6-8 |
| DOI: | 10.1007/s00371-011-0592-9 |
| Kurzbeschreibung (Abstract): | We present a genetic algorithm for approximating densely sampled curves with uniform cubic B-Splines suitable for Combined B-reps. A feature of this representation is altering the continuity property of the B-Spline at any knot, allowing combining freeform curves and polygonal parts within one representation. Naturally there is a trade-off between different approximation properties like accuracy and the number of control points needed. Our algorithm creates very accurate B-Splines with few control points, as shown in Fig. 1. Since the approximation problem is highly nonlinear, we approach it with genetic methods, leading to better results compared to classical gradient based methods. Parallelization and adapted evolution strategies are used to create results very fast. |
| Freie Schlagworte: | Forschungsgruppe Semantic Models, Immersive Systems (SMIS), Business Field: Virtual engineering, Research Area: Semantics in the modeling process, Splines, Approximation, Subdivision surfaces, Genetic algorithms, Parallelization, Combined B-reps |
| Fachbereich(e)/-gebiet(e): | 20 Fachbereich Informatik 20 Fachbereich Informatik > Graphisch-Interaktive Systeme |
| Hinterlegungsdatum: | 12 Nov 2018 11:16 |
| Letzte Änderung: | 04 Feb 2022 12:40 |
| PPN: | |
| Export: | |
| Suche nach Titel in: | TUfind oder in Google |
 |
Frage zum Eintrag |
Optionen (nur für Redakteure)
 |
Redaktionelle Details anzeigen |
Drucken
Impressum