Hello Carsten, thanks for the link. I am not interested in colission detection, it was just stated in the tutorial as an example. I am interested in finding the user selection in somewhat larger range scan datasets of 100000 and more vertices. As I have to process the selection afterwards (selection of the local neighborhood subject to some constraints), it has to be quick. I have another question concerning selection of the local neighborhood. I think the best way to do that is based on a half edge data structure. I have read somewhere, I think it was related to OpenSGPlus, that it is planed to integrate OpenMesh (www.openmesh.org) as a geometry kernel into OpenSG. Therefore I thought to use it in my project as well. But I have now seen that there is already a half edge data structure implementation in OpenSG. Have you made some experience with it and recommend it or would it be better to use OpenMesh instead. I think I need to state further that the data structure is only relevant for polygon selection. I am not interested in other algorithms which come with OpenMesh like mesh-smoothing, etc.
Cheers Christoph -------- Original-Nachricht -------- > Datum: Fri, 14 Mar 2008 15:04:39 -0500 > Von: Carsten Neumann <[EMAIL PROTECTED]> > An: [email protected] > Betreff: Re: [Opensg-users] color picking in opensg > Hello Christoph, > > Christoph John wrote: > > Hey Daniel, > > thanks I just found the tutorial. But I remembered that the OpenSG > tutorial stated somewhere that ray intersection performs quite weak in > presence > of larger geometry. It says it should not be used for collision detection > and things like that as it checks for ray intersection on each object > (polygon? have not checked the source yet). > > Therefore I was interested if there is a faster way to get to the > polygon like color picking. > > There is an tutorial which uses this technique for fast object picking > but not for single polygons. I assume I have to implement it myself :( > > right, the color picking tutorial (Tutorials/30ID_Buffer.cpp) works on > the object level not the polygon level. The IntersectAction uses the > bounding boxes of the objects in the scene to determine which object(s) > are hit, then goes through all polygons to distinguish close objects and > get the hit point and normal etc. > This last part is what makes the operation somewhat slow, as the OpenSG > data structures are optimized for fast rendering not fast intersection > tests. > However, I would not expect color picking on the level of individual > polygons to be very fast either, as that requires a state change for > each polygon which slows the rendering pipeline down. > If you are interested in collision detected, take a look at the CollDet > library (<http://cg.in.tu-clausthal.de/research/colldet/>). > > Hope it helps, > Carsten > > ------------------------------------------------------------------------- > This SF.net email is sponsored by: Microsoft > Defy all challenges. Microsoft(R) Visual Studio 2008. > http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > _______________________________________________ > Opensg-users mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/opensg-users -- Ist Ihr Browser Vista-kompatibel? Jetzt die neuesten Browser-Versionen downloaden: http://www.gmx.net/de/go/browser ------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ _______________________________________________ Opensg-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/opensg-users
