In this paper, we present a user-assisted sketch-based framework to extract hi-level primitives (e.g. columns or staircases) from scanned3D models of an architectural complex. The framework offers a unified level of representation of the hi-level primitives, so that new types of primitives can be easily added as plug-ins to the main engine. Primitives are fitted with a user-assisted procedure: the user suggests the approximate location of the primitive by means of simple mouse gestures, sketched over a rendering of the model. The viewpoint that was selected prior to the sketching is also taken in consideration as hints on the orientation and size of the primitive. The engine performs a GPU assisted fitting and the result is shown in real time to the user. Ad-hoc gestures cause the system to add and fit groups of primitive in one go (e.g. a column complex, or a sequence of windows).
A framework for User-Assisted Sketch-Based Fitting of Geometric Primitives
TARINI, MARCO;
2010-01-01
Abstract
In this paper, we present a user-assisted sketch-based framework to extract hi-level primitives (e.g. columns or staircases) from scanned3D models of an architectural complex. The framework offers a unified level of representation of the hi-level primitives, so that new types of primitives can be easily added as plug-ins to the main engine. Primitives are fitted with a user-assisted procedure: the user suggests the approximate location of the primitive by means of simple mouse gestures, sketched over a rendering of the model. The viewpoint that was selected prior to the sketching is also taken in consideration as hints on the orientation and size of the primitive. The engine performs a GPU assisted fitting and the result is shown in real time to the user. Ad-hoc gestures cause the system to add and fit groups of primitive in one go (e.g. a column complex, or a sequence of windows).File | Dimensione | Formato | |
---|---|---|---|
fitting_primitives.pdf
accesso aperto
Tipologia:
Documento in Pre-print
Licenza:
DRM non definito
Dimensione
4.02 MB
Formato
Adobe PDF
|
4.02 MB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.