You can subscribe to this list here.
| 2005 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
(3) |
Nov
(50) |
Dec
(118) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2006 |
Jan
(93) |
Feb
(17) |
Mar
(6) |
Apr
|
May
(4) |
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
(45) |
Dec
(17) |
| 2007 |
Jan
(23) |
Feb
(33) |
Mar
(4) |
Apr
(1) |
May
|
Jun
|
Jul
(2) |
Aug
(6) |
Sep
(1) |
Oct
|
Nov
(13) |
Dec
(20) |
| 2008 |
Jan
(26) |
Feb
(69) |
Mar
(72) |
Apr
(32) |
May
(35) |
Jun
(12) |
Jul
(99) |
Aug
(52) |
Sep
(15) |
Oct
(49) |
Nov
(12) |
Dec
(26) |
| 2009 |
Jan
(7) |
Feb
|
Mar
(5) |
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
|
From: Paolo C. <pao...@is...> - 2008-08-05 06:41:41
|
skunkwerk wrote: > i tried the hausdorff distance plugin on a number of models, such as > www.akbars.net/20.zip > i noticed two issues: > 1) when the model & a copy of itself in another layer are open, the quadric > texture plugin doesn't seem to reduce models - or at least, the display > isn't updated as the model looks exactly the same. > Probably you have simply forgotten to disable the view of one layer that occludes the simplified one. > 2) the hausdorff distance doesn't really tell you a lot - on one model, i > got the following results after a series of 6 reductions: > > 0.001267 > 0.012462 > 0.035062 > 0.040926 > 0.036652 > 0.027046 > > while on the model mentioned above (20.zip) i got the following: > .99 > 1.7 > 4.7 > 8.8 > 11.9 > 14 > 14.4 > 15.3 > 15.3 > > how can you get some sense of the "quality" of the reduction from looking at > these numbers? any ideas? > Read these papers. accurately. http://vcg.isti.cnr.it/publications/papers/metro.pdf http://citeseer.ist.psu.edu/cignoni97comparison.html -- Paolo Cignoni -- Senior Researcher Visual Computing Laboratory - ISTI - CNR http://vcg.isti.cnr.it/~cignoni ISTI - CNR Via Moruzzi 1, 56124 Pisa ITALY |
|
From: Paolo C. <pao...@is...> - 2008-08-05 06:35:53
|
skunkwerk wrote: > thank you Paolo, > > 1) i don't know what i was doing wrong before, but I tried the layers > again, and it worked fine. the hausdorff distance plugin gave me the > following output when I ran it on 2 identical meshes: > max:65, mean:0.026, RMS:0.99 > doing more samples would likely bring these values closer to 0, right?... > would the best way to interpret these values be to compare them to the > radius of the mesh's bounding box? > > yes if the two meshes are perfectly equal. -- Paolo Cignoni -- Senior Researcher Visual Computing Laboratory - ISTI - CNR http://vcg.isti.cnr.it/~cignoni ISTI - CNR Via Moruzzi 1, 56124 Pisa ITALY |
|
From: skunkwerk <sku...@gm...> - 2008-08-05 03:36:46
|
i tried the hausdorff distance plugin on a number of models, such as www.akbars.net/20.zip i noticed two issues: 1) when the model & a copy of itself in another layer are open, the quadric texture plugin doesn't seem to reduce models - or at least, the display isn't updated as the model looks exactly the same. 2) the hausdorff distance doesn't really tell you a lot - on one model, i got the following results after a series of 6 reductions: 0.001267 0.012462 0.035062 0.040926 0.036652 0.027046 while on the model mentioned above (20.zip) i got the following: .99 1.7 4.7 8.8 11.9 14 14.4 15.3 15.3 how can you get some sense of the "quality" of the reduction from looking at these numbers? any ideas? On Mon, Aug 4, 2008 at 5:47 PM, skunkwerk <sku...@gm...> wrote: > thank you Paolo, > > 1) i don't know what i was doing wrong before, but I tried the layers > again, and it worked fine. the hausdorff distance plugin gave me the > following output when I ran it on 2 identical meshes: > max:65, mean:0.026, RMS:0.99 > doing more samples would likely bring these values closer to 0, right?... > would the best way to interpret these values be to compare them to the > radius of the mesh's bounding box? > > 2) i tried the minerva ply file on a fresh checkout of meshlab, and it > froze my whole machine - so yes, it'd be good if someone else could try this > out > > 3) they're some comments in Italian as well inside refine.h, i'll wait for > your update > > 4) my question about finding the number of "leaves" was in relation to the > discussion of the model here: www.akbars.net/27.zip > to which you said: "one of the tree has approx 26k faces in the leaves. All > the leaves has no internal vertex..." > > 5) it seems like the 'border' in meshlab is the border of a simplicial > complex... when I thought of boundary I was thinking of the outermost faces > of a model. but i guess the real issue is that simplification with > 'preserve borders' sometimes works well, and sometimes not - and i'm trying > to figure out a way to figure out when to use it > > regards, > imran > > > On Mon, Aug 4, 2008 at 2:31 PM, Paolo Cignoni <pao...@is...>wrote: > >> skunkwerk wrote: >> >> thanks Paolo, >> it's nice to have you back. i didn't feel the wiki page answered my >> questions: >> 1) what's the definition of a border & boundary within meshlab, and are all >> non-manifold faces bordering faces? >> >> >> I have tried to improve the doc. >> >> >> http://vcg.sourceforge.net/index.php/Tutorial#Boundary_relations_and_adjacency >> >> It is nice to have someone that actually tests our scarce documentation :) >> >> >> 2) do the number of leaves correspond to the number of border faces, or >> what? >> >> >> I have not understood what do you mean with leaves? what leaves? what >> tree are you referring to? I have lost the context. >> >> while these are from other threads, i don't want to send 5 emails to the >> list: >> 3) the same file you sent me to test on the texture filter froze my entire >> machine & meshlab - did you not encounter any issues with it? was it not >> working with the 'copied tiles' option or with 'resampled'? >> >> >> yes it does not works with your filter. It is a typical dirty mesh >> against which algorithms should be made robust. >> (the mesh we are talking about is the old textured minerva 250 face and >> two textures) >> >> 4) i tried loading 2 layers into meshlab to use with the hausdorff distance >> plugin, but i don't think the layers are working. any suggestions? >> >> >> I have tried on different architecture and layers seems to be correctly >> working. >> Can some one be able to replicate this issue? >> >> 5) trying to clip triangles with refine.h: you said i need to define which >> edges of the triangle need to be split - but how do I define the points it >> should be split at? i tried doing a search for 'subdivision' in the >> codebase but couldn't find much - could someone point me to the specific >> file that does some clipping? >> >> >> I will expand the comment in the refine.h file >> look for the next commits. >> p. >> >> >> regards, >> imran >> >> On Sun, Aug 3, 2008 at 1:26 AM, Paolo Cignoni <pao...@is...> <pao...@is...>wrote: >> >> >> >> skunkwerk wrote: >> >> >> Paolo, may I ask how you determined the number of leaves in the model? >> >> >> The >> >> >> .dae had a maximum of 2,766 triangles in one "node" >> >> sorry for all the questions >> >> imran >> >> On Mon, Jul 21, 2008 at 6:10 AM, Paolo Cignoni < >> >> >> pao...@is...>wrote: >> >> >> I have added some notes on the border flags in the vcg wikihttp://vcg.sourceforge.net/index.php/Flags >> p. >> >> >> >> -- >> Paolo Cignoni -- Senior Researcher >> Visual Computing Laboratory - ISTI - CNR >> http://vcg.isti.cnr.it/~cignoni <http://vcg.isti.cnr.it/%7Ecignoni> <http://vcg.isti.cnr.it/%7Ecignoni> <http://vcg.isti.cnr.it/%7Ecignoni> >> >> >> ISTI - CNR >> Via Moruzzi 1, >> 56124 Pisa >> ITALY >> >> >> ------------------------------------------------------------------------- >> This SF.Net email is sponsored by the Moblin Your Move Developer's >> challenge >> Build the coolest Linux based applications with Moblin SDK & win great >> prizes >> Grand prize is a trip for two to an Open Source event anywhere in the worldhttp://moblin-contest.org/redirect.php?banner_id=100&url=/ >> _______________________________________________ >> Meshlab-devel mailing lis...@li...https://lists.sourceforge.net/lists/listinfo/meshlab-devel >> >> ------------------------------ >> >> ------------------------------------------------------------------------- >> This SF.Net email is sponsored by the Moblin Your Move Developer's challenge >> Build the coolest Linux based applications with Moblin SDK & win great prizes >> Grand prize is a trip for two to an Open Source event anywhere in the worldhttp://moblin-contest.org/redirect.php?banner_id=100&url=/ >> >> ------------------------------ >> >> _______________________________________________ >> Meshlab-devel mailing lis...@li...https://lists.sourceforge.net/lists/listinfo/meshlab-devel >> >> >> >> -- >> Paolo Cignoni -- Senior Researcher >> Visual Computing Laboratory - ISTI - CNR http://vcg.isti.cnr.it/~cignoni <http://vcg.isti.cnr.it/%7Ecignoni> >> >> ISTI - CNR >> Via Moruzzi 1, >> 56124 Pisa >> ITALY >> >> >> ------------------------------------------------------------------------- >> This SF.Net email is sponsored by the Moblin Your Move Developer's >> challenge >> Build the coolest Linux based applications with Moblin SDK & win great >> prizes >> Grand prize is a trip for two to an Open Source event anywhere in the >> world >> http://moblin-contest.org/redirect.php?banner_id=100&url=/ >> _______________________________________________ >> Meshlab-devel mailing list >> Mes...@li... >> https://lists.sourceforge.net/lists/listinfo/meshlab-devel >> >> > |
|
From: skunkwerk <sku...@gm...> - 2008-08-05 00:48:00
|
thank you Paolo, 1) i don't know what i was doing wrong before, but I tried the layers again, and it worked fine. the hausdorff distance plugin gave me the following output when I ran it on 2 identical meshes: max:65, mean:0.026, RMS:0.99 doing more samples would likely bring these values closer to 0, right?... would the best way to interpret these values be to compare them to the radius of the mesh's bounding box? 2) i tried the minerva ply file on a fresh checkout of meshlab, and it froze my whole machine - so yes, it'd be good if someone else could try this out 3) they're some comments in Italian as well inside refine.h, i'll wait for your update 4) my question about finding the number of "leaves" was in relation to the discussion of the model here: www.akbars.net/27.zip to which you said: "one of the tree has approx 26k faces in the leaves. All the leaves has no internal vertex..." 5) it seems like the 'border' in meshlab is the border of a simplicial complex... when I thought of boundary I was thinking of the outermost faces of a model. but i guess the real issue is that simplification with 'preserve borders' sometimes works well, and sometimes not - and i'm trying to figure out a way to figure out when to use it regards, imran On Mon, Aug 4, 2008 at 2:31 PM, Paolo Cignoni <pao...@is...>wrote: > skunkwerk wrote: > > thanks Paolo, > it's nice to have you back. i didn't feel the wiki page answered my > questions: > 1) what's the definition of a border & boundary within meshlab, and are all > non-manifold faces bordering faces? > > > I have tried to improve the doc. > > > http://vcg.sourceforge.net/index.php/Tutorial#Boundary_relations_and_adjacency > > It is nice to have someone that actually tests our scarce documentation :) > > > 2) do the number of leaves correspond to the number of border faces, or > what? > > > I have not understood what do you mean with leaves? what leaves? what tree > are you referring to? I have lost the context. > > while these are from other threads, i don't want to send 5 emails to the > list: > 3) the same file you sent me to test on the texture filter froze my entire > machine & meshlab - did you not encounter any issues with it? was it not > working with the 'copied tiles' option or with 'resampled'? > > > yes it does not works with your filter. It is a typical dirty mesh > against which algorithms should be made robust. > (the mesh we are talking about is the old textured minerva 250 face and two > textures) > > 4) i tried loading 2 layers into meshlab to use with the hausdorff distance > plugin, but i don't think the layers are working. any suggestions? > > > I have tried on different architecture and layers seems to be correctly > working. > Can some one be able to replicate this issue? > > 5) trying to clip triangles with refine.h: you said i need to define which > edges of the triangle need to be split - but how do I define the points it > should be split at? i tried doing a search for 'subdivision' in the > codebase but couldn't find much - could someone point me to the specific > file that does some clipping? > > > I will expand the comment in the refine.h file > look for the next commits. > p. > > > regards, > imran > > On Sun, Aug 3, 2008 at 1:26 AM, Paolo Cignoni <pao...@is...> <pao...@is...>wrote: > > > > skunkwerk wrote: > > > Paolo, may I ask how you determined the number of leaves in the model? > > > The > > > .dae had a maximum of 2,766 triangles in one "node" > > sorry for all the questions > > imran > > On Mon, Jul 21, 2008 at 6:10 AM, Paolo Cignoni < > > > pao...@is...>wrote: > > > I have added some notes on the border flags in the vcg wikihttp://vcg.sourceforge.net/index.php/Flags > p. > > > > -- > Paolo Cignoni -- Senior Researcher > Visual Computing Laboratory - ISTI - CNR > http://vcg.isti.cnr.it/~cignoni <http://vcg.isti.cnr.it/%7Ecignoni> <http://vcg.isti.cnr.it/%7Ecignoni> <http://vcg.isti.cnr.it/%7Ecignoni> > > ISTI - CNR > Via Moruzzi 1, > 56124 Pisa > ITALY > > > ------------------------------------------------------------------------- > This SF.Net email is sponsored by the Moblin Your Move Developer's > challenge > Build the coolest Linux based applications with Moblin SDK & win great > prizes > Grand prize is a trip for two to an Open Source event anywhere in the worldhttp://moblin-contest.org/redirect.php?banner_id=100&url=/ > _______________________________________________ > Meshlab-devel mailing lis...@li...https://lists.sourceforge.net/lists/listinfo/meshlab-devel > > ------------------------------ > > ------------------------------------------------------------------------- > This SF.Net email is sponsored by the Moblin Your Move Developer's challenge > Build the coolest Linux based applications with Moblin SDK & win great prizes > Grand prize is a trip for two to an Open Source event anywhere in the worldhttp://moblin-contest.org/redirect.php?banner_id=100&url=/ > > ------------------------------ > > _______________________________________________ > Meshlab-devel mailing lis...@li...https://lists.sourceforge.net/lists/listinfo/meshlab-devel > > > > -- > Paolo Cignoni -- Senior Researcher > Visual Computing Laboratory - ISTI - CNR http://vcg.isti.cnr.it/~cignoni <http://vcg.isti.cnr.it/%7Ecignoni> > > ISTI - CNR > Via Moruzzi 1, > 56124 Pisa > ITALY > > > ------------------------------------------------------------------------- > This SF.Net email is sponsored by the Moblin Your Move Developer's > challenge > Build the coolest Linux based applications with Moblin SDK & win great > prizes > Grand prize is a trip for two to an Open Source event anywhere in the world > http://moblin-contest.org/redirect.php?banner_id=100&url=/ > _______________________________________________ > Meshlab-devel mailing list > Mes...@li... > https://lists.sourceforge.net/lists/listinfo/meshlab-devel > > |
|
From: Paolo C. <pao...@is...> - 2008-08-04 21:32:20
|
skunkwerk wrote: > thanks Paolo, > it's nice to have you back. i didn't feel the wiki page answered my > questions: > 1) what's the definition of a border & boundary within meshlab, and are all > non-manifold faces bordering faces? > I have tried to improve the doc. http://vcg.sourceforge.net/index.php/Tutorial#Boundary_relations_and_adjacency It is nice to have someone that actually tests our scarce documentation :) > 2) do the number of leaves correspond to the number of border faces, or > what? > I have not understood what do you mean with leaves? what leaves? what tree are you referring to? I have lost the context. > while these are from other threads, i don't want to send 5 emails to the > list: > 3) the same file you sent me to test on the texture filter froze my entire > machine & meshlab - did you not encounter any issues with it? was it not > working with the 'copied tiles' option or with 'resampled'? > yes it does not works with your filter. It is a typical dirty mesh against which algorithms should be made robust. (the mesh we are talking about is the old textured minerva 250 face and two textures) > 4) i tried loading 2 layers into meshlab to use with the hausdorff distance > plugin, but i don't think the layers are working. any suggestions? > I have tried on different architecture and layers seems to be correctly working. Can some one be able to replicate this issue? > 5) trying to clip triangles with refine.h: you said i need to define which > edges of the triangle need to be split - but how do I define the points it > should be split at? i tried doing a search for 'subdivision' in the > codebase but couldn't find much - could someone point me to the specific > file that does some clipping? > I will expand the comment in the refine.h file look for the next commits. p. > regards, > imran > > On Sun, Aug 3, 2008 at 1:26 AM, Paolo Cignoni <pao...@is...>wrote: > > >> skunkwerk wrote: >> >>> Paolo, may I ask how you determined the number of leaves in the model? >>> >> The >> >>> .dae had a maximum of 2,766 triangles in one "node" >>> >>> sorry for all the questions >>> >>> imran >>> >>> On Mon, Jul 21, 2008 at 6:10 AM, Paolo Cignoni < >>> >> pao...@is...>wrote: >> >> I have added some notes on the border flags in the vcg wiki >> http://vcg.sourceforge.net/index.php/Flags >> p. >> >> >> >> -- >> Paolo Cignoni -- Senior Researcher >> Visual Computing Laboratory - ISTI - CNR >> http://vcg.isti.cnr.it/~cignoni <http://vcg.isti.cnr.it/%7Ecignoni> >> >> ISTI - CNR >> Via Moruzzi 1, >> 56124 Pisa >> ITALY >> >> >> ------------------------------------------------------------------------- >> This SF.Net email is sponsored by the Moblin Your Move Developer's >> challenge >> Build the coolest Linux based applications with Moblin SDK & win great >> prizes >> Grand prize is a trip for two to an Open Source event anywhere in the world >> http://moblin-contest.org/redirect.php?banner_id=100&url=/ >> _______________________________________________ >> Meshlab-devel mailing list >> Mes...@li... >> https://lists.sourceforge.net/lists/listinfo/meshlab-devel >> >> > > > ------------------------------------------------------------------------ > > ------------------------------------------------------------------------- > This SF.Net email is sponsored by the Moblin Your Move Developer's challenge > Build the coolest Linux based applications with Moblin SDK & win great prizes > Grand prize is a trip for two to an Open Source event anywhere in the world > http://moblin-contest.org/redirect.php?banner_id=100&url=/ > > ------------------------------------------------------------------------ > > _______________________________________________ > Meshlab-devel mailing list > Mes...@li... > https://lists.sourceforge.net/lists/listinfo/meshlab-devel > -- Paolo Cignoni -- Senior Researcher Visual Computing Laboratory - ISTI - CNR http://vcg.isti.cnr.it/~cignoni ISTI - CNR Via Moruzzi 1, 56124 Pisa ITALY |
|
From: fabio g. <fab...@gm...> - 2008-08-04 19:39:22
|
Hi,
you can use the GetClosestFace function in vcg/trimesh/closest.h.
Include vcg::face::EdgePlane as component of your face type and call
vcg::UpdateEdge<MeshModel>::Set(..) which sets the values of EdgePlane
for each face.
If you don't do this, there quantities will be computed on demand
(multiple times) and your code will
be considerably slower.
Look at the project apps/metro/ (file sampling.h) to see a complete example.
cheers,
Fabio
> Hi all,
>
> i've started working on the "re-topology" plugin (just committed a
> *very* initial version),
> and i've got a question on how to use (correctly) the Uniform Grid object.
>
> My goal is to get the closest point of a mesh from a given spatial
> coord (not a vertex of
> the mesh structure, just a free point over its surface),
> and then use that point to build the retopologized mesh.
>
> In order to do this, i've looked for some spatial indexing structures,
> and found the
> one that is called "GridStaticPtr", whitch seems to be exaclty what i needed.
>
>
> So, once initialized the Grid object ("unifGrid"):
>
>
> meshModel->updateDataMask(MeshModel::MM_FACEMARK);
> unifGrid.Set(mesh->face.begin(), mesh->face.end());
> markerFunctor.SetMesh(mesh);
>
>
> i use this code to get the closest point:
>
>
> float dist_upper_bound =mesh->bbox.Diag()/50;
> float dist = dist_upper_bound;
> const CMeshO::CoordType &startPt= toCheck;
>
> CMeshO::FaceType *nearestF=0;
> vcg::face::PointDistanceBaseFunctor PDistFunct;
>
> nearestF = unifGrid.GetClosest(PDistFunct, markerFunctor, startPt,
> dist_upper_bound, dist, closestPt);
>
>
> Here i expect to get in "closestPt" the mesh surface point nearest to
> "startPt", but this does
> not happen for every starting point passed (expecially on hi-res meshes, the
> returned point is not just the nonclosest, but it's also not a point
> of the mesh).
>
>
> Well, my question is: am i using "unifGrid" correctly?
> And, if not, any suggestion on how to get the closest mesh point from
> a given coord (using uniform grid or something else)?
>
> Tanks in advance,
> D.
>
> -------------------------------------------------------------------------
> This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
> Build the coolest Linux based applications with Moblin SDK & win great prizes
> Grand prize is a trip for two to an Open Source event anywhere in the world
> http://moblin-contest.org/redirect.php?banner_id=100&url=/
> _______________________________________________
> Meshlab-devel mailing list
> Mes...@li...
> https://lists.sourceforge.net/lists/listinfo/meshlab-devel
>
>
|
|
From: Daniele B. <ele...@gm...> - 2008-08-04 18:02:24
|
Hi all,
i've started working on the "re-topology" plugin (just committed a
*very* initial version),
and i've got a question on how to use (correctly) the Uniform Grid object.
My goal is to get the closest point of a mesh from a given spatial
coord (not a vertex of
the mesh structure, just a free point over its surface),
and then use that point to build the retopologized mesh.
In order to do this, i've looked for some spatial indexing structures,
and found the
one that is called "GridStaticPtr", whitch seems to be exaclty what i needed.
So, once initialized the Grid object ("unifGrid"):
meshModel->updateDataMask(MeshModel::MM_FACEMARK);
unifGrid.Set(mesh->face.begin(), mesh->face.end());
markerFunctor.SetMesh(mesh);
i use this code to get the closest point:
float dist_upper_bound =mesh->bbox.Diag()/50;
float dist = dist_upper_bound;
const CMeshO::CoordType &startPt= toCheck;
CMeshO::FaceType *nearestF=0;
vcg::face::PointDistanceBaseFunctor PDistFunct;
nearestF = unifGrid.GetClosest(PDistFunct, markerFunctor, startPt,
dist_upper_bound, dist, closestPt);
Here i expect to get in "closestPt" the mesh surface point nearest to
"startPt", but this does
not happen for every starting point passed (expecially on hi-res meshes, the
returned point is not just the nonclosest, but it's also not a point
of the mesh).
Well, my question is: am i using "unifGrid" correctly?
And, if not, any suggestion on how to get the closest mesh point from
a given coord (using uniform grid or something else)?
Tanks in advance,
D.
|
|
From: skunkwerk <sku...@gm...> - 2008-08-04 05:09:04
|
problem seems to have been solved - i needed to compact the face vector On Sun, Aug 3, 2008 at 10:01 PM, skunkwerk <sku...@gm...> wrote: > yeah, sorry i was getting sloppy. > > i've moved the code around, so now the error is similar, but the following: > meshlab: ../../../../vcglib/vcg/complex/trimesh/allocate.h:513: static void > vcg::tri::Allocator<AllocateMeshType>::CompactFaceVector(AllocateMeshType&) > [with AllocateMeshType = CMeshO]: Assertion `(int)pos==m.fn' failed. > > the only places I'm touching the mesh in my code are lines like the > following: > > const typename MESHTYPE::FaceType& fp = m.face[plane]; > typename MESHTYPE::VertexType::NormalType r = fp.cN(); > r.Normalize(); > typename MESHTYPE::VertexType v = m.vert[fp.V(0)-&(*m.vert.begin())]; > D = -r[0]*v.P()[0] - r[1]*v.P()[1] -r[2]*v.P()[2];//calculate d for the > plane > > do i need to be using const accessors or something? > > imran > > > On Sun, Aug 3, 2008 at 12:39 AM, Paolo Cignoni <pao...@is...>wrote: > >> Hi imran, >> >> from the MeshLab wiki >> >> http://meshlab.sourceforge.net/wiki/index.php/Writing_Code >> >> ... >> when you report bugs please adhere to the suggestion in >> http://www.chiark.greenend.org.uk/~sgtatham/bugs.html<http://www.chiark.greenend.org.uk/%7Esgtatham/bugs.html> >> (How to Report Bugs Effectively) >> ... >> >> so please elaborate a bit.... >> p. >> >> >> skunkwerk wrote: >> >> i'm getting this error: >> meshlab: ../../../vcglib/vcg/complex/trimesh/update/flag.h:339: static >> void >> vcg::tri::UpdateFlags<UpdateMeshType>::FaceBorderFromNone(UpdateMeshType&) >> [with UpdateMeshType = CMeshO]: Assertion `p==e.end()' failed. >> >> from a call to GLA()->mm()->updateDataMask(req); >> the thing is, i haven't changed the mesh at all - I've just accessed its >> faces and vertices. so i'm not sure why this is happening.. >> any ideas? >> >> thanks, >> imran >> >> ------------------------------ >> >> ------------------------------------------------------------------------- >> This SF.Net email is sponsored by the Moblin Your Move Developer's challenge >> Build the coolest Linux based applications with Moblin SDK & win great prizes >> Grand prize is a trip for two to an Open Source event anywhere in the worldhttp://moblin-contest.org/redirect.php?banner_id=100&url=/ >> >> ------------------------------ >> >> _______________________________________________ >> Meshlab-devel mailing lis...@li...https://lists.sourceforge.net/lists/listinfo/meshlab-devel >> >> >> >> -- >> Paolo Cignoni -- Senior Researcher >> Visual Computing Laboratory - ISTI - CNR http://vcg.isti.cnr.it/~cignoni <http://vcg.isti.cnr.it/%7Ecignoni> >> >> ISTI - CNR >> Via Moruzzi 1, >> 56124 Pisa >> ITALY >> >> >> ------------------------------------------------------------------------- >> This SF.Net email is sponsored by the Moblin Your Move Developer's >> challenge >> Build the coolest Linux based applications with Moblin SDK & win great >> prizes >> Grand prize is a trip for two to an Open Source event anywhere in the >> world >> http://moblin-contest.org/redirect.php?banner_id=100&url=/ >> _______________________________________________ >> Meshlab-devel mailing list >> Mes...@li... >> https://lists.sourceforge.net/lists/listinfo/meshlab-devel >> >> > |
|
From: skunkwerk <sku...@gm...> - 2008-08-04 05:01:12
|
yeah, sorry i was getting sloppy. i've moved the code around, so now the error is similar, but the following: meshlab: ../../../../vcglib/vcg/complex/trimesh/allocate.h:513: static void vcg::tri::Allocator<AllocateMeshType>::CompactFaceVector(AllocateMeshType&) [with AllocateMeshType = CMeshO]: Assertion `(int)pos==m.fn' failed. the only places I'm touching the mesh in my code are lines like the following: const typename MESHTYPE::FaceType& fp = m.face[plane]; typename MESHTYPE::VertexType::NormalType r = fp.cN(); r.Normalize(); typename MESHTYPE::VertexType v = m.vert[fp.V(0)-&(*m.vert.begin())]; D = -r[0]*v.P()[0] - r[1]*v.P()[1] -r[2]*v.P()[2];//calculate d for the plane do i need to be using const accessors or something? imran On Sun, Aug 3, 2008 at 12:39 AM, Paolo Cignoni <pao...@is...>wrote: > Hi imran, > > from the MeshLab wiki > > http://meshlab.sourceforge.net/wiki/index.php/Writing_Code > > ... > when you report bugs please adhere to the suggestion in > http://www.chiark.greenend.org.uk/~sgtatham/bugs.html<http://www.chiark.greenend.org.uk/%7Esgtatham/bugs.html> > (How to Report Bugs Effectively) > ... > > so please elaborate a bit.... > p. > > > skunkwerk wrote: > > i'm getting this error: > meshlab: ../../../vcglib/vcg/complex/trimesh/update/flag.h:339: static void > vcg::tri::UpdateFlags<UpdateMeshType>::FaceBorderFromNone(UpdateMeshType&) > [with UpdateMeshType = CMeshO]: Assertion `p==e.end()' failed. > > from a call to GLA()->mm()->updateDataMask(req); > the thing is, i haven't changed the mesh at all - I've just accessed its > faces and vertices. so i'm not sure why this is happening.. > any ideas? > > thanks, > imran > > ------------------------------ > > ------------------------------------------------------------------------- > This SF.Net email is sponsored by the Moblin Your Move Developer's challenge > Build the coolest Linux based applications with Moblin SDK & win great prizes > Grand prize is a trip for two to an Open Source event anywhere in the worldhttp://moblin-contest.org/redirect.php?banner_id=100&url=/ > > ------------------------------ > > _______________________________________________ > Meshlab-devel mailing lis...@li...https://lists.sourceforge.net/lists/listinfo/meshlab-devel > > > > -- > Paolo Cignoni -- Senior Researcher > Visual Computing Laboratory - ISTI - CNR http://vcg.isti.cnr.it/~cignoni <http://vcg.isti.cnr.it/%7Ecignoni> > > ISTI - CNR > Via Moruzzi 1, > 56124 Pisa > ITALY > > > ------------------------------------------------------------------------- > This SF.Net email is sponsored by the Moblin Your Move Developer's > challenge > Build the coolest Linux based applications with Moblin SDK & win great > prizes > Grand prize is a trip for two to an Open Source event anywhere in the world > http://moblin-contest.org/redirect.php?banner_id=100&url=/ > _______________________________________________ > Meshlab-devel mailing list > Mes...@li... > https://lists.sourceforge.net/lists/listinfo/meshlab-devel > > |
|
From: skunkwerk <sku...@gm...> - 2008-08-04 04:41:58
|
thanks Paolo, it's nice to have you back. i didn't feel the wiki page answered my questions: 1) what's the definition of a border & boundary within meshlab, and are all non-manifold faces bordering faces? 2) do the number of leaves correspond to the number of border faces, or what? while these are from other threads, i don't want to send 5 emails to the list: 3) the same file you sent me to test on the texture filter froze my entire machine & meshlab - did you not encounter any issues with it? was it not working with the 'copied tiles' option or with 'resampled'? 4) i tried loading 2 layers into meshlab to use with the hausdorff distance plugin, but i don't think the layers are working. any suggestions? 5) trying to clip triangles with refine.h: you said i need to define which edges of the triangle need to be split - but how do I define the points it should be split at? i tried doing a search for 'subdivision' in the codebase but couldn't find much - could someone point me to the specific file that does some clipping? regards, imran On Sun, Aug 3, 2008 at 1:26 AM, Paolo Cignoni <pao...@is...>wrote: > skunkwerk wrote: > > Paolo, may I ask how you determined the number of leaves in the model? > The > > .dae had a maximum of 2,766 triangles in one "node" > > > > sorry for all the questions > > > > imran > > > > On Mon, Jul 21, 2008 at 6:10 AM, Paolo Cignoni < > pao...@is...>wrote: > > > I have added some notes on the border flags in the vcg wiki > http://vcg.sourceforge.net/index.php/Flags > p. > > > > -- > Paolo Cignoni -- Senior Researcher > Visual Computing Laboratory - ISTI - CNR > http://vcg.isti.cnr.it/~cignoni <http://vcg.isti.cnr.it/%7Ecignoni> > > ISTI - CNR > Via Moruzzi 1, > 56124 Pisa > ITALY > > > ------------------------------------------------------------------------- > This SF.Net email is sponsored by the Moblin Your Move Developer's > challenge > Build the coolest Linux based applications with Moblin SDK & win great > prizes > Grand prize is a trip for two to an Open Source event anywhere in the world > http://moblin-contest.org/redirect.php?banner_id=100&url=/ > _______________________________________________ > Meshlab-devel mailing list > Mes...@li... > https://lists.sourceforge.net/lists/listinfo/meshlab-devel > |
|
From: Paolo C. <pao...@is...> - 2008-08-03 08:26:53
|
skunkwerk wrote: > Paolo, may I ask how you determined the number of leaves in the model? The > .dae had a maximum of 2,766 triangles in one "node" > > sorry for all the questions > > imran > > On Mon, Jul 21, 2008 at 6:10 AM, Paolo Cignoni <pao...@is...>wrote: > I have added some notes on the border flags in the vcg wiki http://vcg.sourceforge.net/index.php/Flags p. -- Paolo Cignoni -- Senior Researcher Visual Computing Laboratory - ISTI - CNR http://vcg.isti.cnr.it/~cignoni ISTI - CNR Via Moruzzi 1, 56124 Pisa ITALY |
|
From: Paolo C. <pao...@is...> - 2008-08-03 07:40:35
|
Edoardo Pocci wrote: > Ciao a tutti,mi servirebbe un aiuto perchè non riesco a trovar euna soluzione,è possibile trovare una rotazione che permette, partendo da un piano di arrivare su un'altro piano?Va bene anche se la rotazione avviene sul centro del piano. > Se non sbaglio si dovrebbero trovare 3 rotazioni sui 3 assi.Ciao e grazie > Edoardo Pocci > writing in English is appreciated :) Could you try to elaborate a bit your needs? Why do you need this in your (edit slice) plugin? IMHO you should just need to translate current trackball transformation into object space (something that the old edit plugins already did). Perhaps i am missing something. Could you try to better explain where and why do you need to transform a plane into another plane? Cheers p. -- Paolo Cignoni -- Senior Researcher Visual Computing Laboratory - ISTI - CNR http://vcg.isti.cnr.it/~cignoni ISTI - CNR Via Moruzzi 1, 56124 Pisa ITALY |
|
From: Paolo C. <pao...@is...> - 2008-08-03 07:40:34
|
Hi imran, from the MeshLab wiki http://meshlab.sourceforge.net/wiki/index.php/Writing_Code ... when you report bugs please adhere to the suggestion in http://www.chiark.greenend.org.uk/~sgtatham/bugs.html (How to Report Bugs Effectively) ... so please elaborate a bit.... p. skunkwerk wrote: > i'm getting this error: > meshlab: ../../../vcglib/vcg/complex/trimesh/update/flag.h:339: static > void > vcg::tri::UpdateFlags<UpdateMeshType>::FaceBorderFromNone(UpdateMeshType&) > [with UpdateMeshType = CMeshO]: Assertion `p==e.end()' failed. > > from a call to GLA()->mm()->updateDataMask(req); > the thing is, i haven't changed the mesh at all - I've just accessed > its faces and vertices. so i'm not sure why this is happening.. > any ideas? > > thanks, > imran > ------------------------------------------------------------------------ > > ------------------------------------------------------------------------- > This SF.Net email is sponsored by the Moblin Your Move Developer's challenge > Build the coolest Linux based applications with Moblin SDK & win great prizes > Grand prize is a trip for two to an Open Source event anywhere in the world > http://moblin-contest.org/redirect.php?banner_id=100&url=/ > > ------------------------------------------------------------------------ > > _______________________________________________ > Meshlab-devel mailing list > Mes...@li... > https://lists.sourceforge.net/lists/listinfo/meshlab-devel > -- Paolo Cignoni -- Senior Researcher Visual Computing Laboratory - ISTI - CNR http://vcg.isti.cnr.it/~cignoni ISTI - CNR Via Moruzzi 1, 56124 Pisa ITALY |
|
From: Paolo C. <pao...@is...> - 2008-08-03 07:40:30
|
skunkwerk wrote: > thanks Fabio, i'll try it out. I was also looking for where in the > code it splits triangles that intersect the plane you're dividing with > - I remember Paolo said something about how the refine.h file can clip > triangles; do you know of any code that uses this? > yes refine can clip single triangles. You have just to define a predicate that says for each given edge if it has to be splitted or not. A lot of code use it, all the subdivision surface code use it (look at the corresponding meshlab filter). p. > is Paolo on vacation? > yes, but I being still at home I still read email and compile meshlab... :( p. > imran > > On Thu, Jul 31, 2008 at 1:47 PM, fabio ganovelli > <fab...@gm... <mailto:fab...@gm...>> wrote: > > Hi, > I think you overseen the project metro? (vcglib/apps/metro) > In the file sampling.h of that project you can read the declaration: > > typedef Octree<FaceType, typename MetroMesh::ScalarType > > MetroMeshOctree; > > and some lines later the definition: > > MetroMeshOctree oS2; > > and finally the population: > > if(Flags & SamplingFlags::USE_OCTREE) > oS2.Set(S2.face.begin(),S2.face.end()); > > which (just like the other SI structure) is done with the function Set > which takes > the iterators to the first and the last element to store. > > However, you can also find the implementation in the file > vcglib/space/octree.h > (line 231) > > /*! > * Populate the octree > */ > template < class OBJECT_ITERATOR > > void Set(const OBJECT_ITERATOR & bObj, const OBJECT_ITERATOR & > eObj /*, vcg::CallBackPos *callback=NULL*/) > > with an English comment :) > > cheers, > Fabio > > > > > thanks Fabio, > > i looked at the files you mentioned... an octree would work fine > > for my needs, if i could figure out how to use the code in meshlab - > > is there a function somewhere to actually populate an octree from a > > given model? the initialize function in octree_template.h doesn't > > seem to do much, and the comments are in Italian! > > > > imran > > > > On Thu, Jul 31, 2008 at 2:41 AM, fabio ganovelli > > <fab...@gm... <mailto:fab...@gm...> > <mailto:fab...@gm... > <mailto:fab...@gm...>>> wrote: > > > > The files in /vcglib/vcg/space/index implement a few spatial > indexing > > data structures > > and algorithms (regular grid, octree, spatial hashing..) > with a common > > interface. > > This means that you have function to get the closest k > elements in the > > spatial index > > without caring if it is a grid or an octree. See > > vcg/complex/trimesh/closest.h to get the picture of how it is > > done. Even > > better, look at the project metro in vcglib/apps/metro which > computes > > "point to mesh" distance with (optionally) all the spatial > > indexing data > > structures. > > > > At the present the spatial indexing ds include vicinity > tests and > > (ordered) ray intersection test, but no z sorting and no > generic BSP. > > > > cheers, > > Fabio > > > > > > > I'm looking to create a binary space partitioning (BSP) > tree in > > > meshlab, to use when rendering on another platform for correct > > > z-sorting. I noticed two files already present: octree.h in > > > /vcglib/vcg/space/index, and aabb_binary_tree.h in > > > /vcglib/vcg/space/index (written by m_di_benedetto > > > <http://sourceforge.net/users/m_di_benedetto/>)? Does > anyone know > > > what these files do, and if they could be of any use for > what i'm > > > trying to do? > > > > > > thanks > > > imran > > > > > > ------------------------------------------------------------------------ > > > > > > > > > ------------------------------------------------------------------------- > > > This SF.Net email is sponsored by the Moblin Your Move > > Developer's challenge > > > Build the coolest Linux based applications with Moblin SDK > & win > > great prizes > > > Grand prize is a trip for two to an Open Source event anywhere > > in the world > > > http://moblin-contest.org/redirect.php?banner_id=100&url=/ > <http://moblin-contest.org/redirect.php?banner_id=100&url=/> > > <http://moblin-contest.org/redirect.php?banner_id=100&url=/ > <http://moblin-contest.org/redirect.php?banner_id=100&url=/>> > > > > > > ------------------------------------------------------------------------ > > > > > > _______________________________________________ > > > Meshlab-devel mailing list > > > Mes...@li... > <mailto:Mes...@li...> > > <mailto:Mes...@li... > <mailto:Mes...@li...>> > > > https://lists.sourceforge.net/lists/listinfo/meshlab-devel > > > > > > > > > > ------------------------------------------------------------------------- > > This SF.Net email is sponsored by the Moblin Your Move > Developer's > > challenge > > Build the coolest Linux based applications with Moblin SDK & win > > great prizes > > Grand prize is a trip for two to an Open Source event > anywhere in > > the world > > http://moblin-contest.org/redirect.php?banner_id=100&url=/ > <http://moblin-contest.org/redirect.php?banner_id=100&url=/> > > <http://moblin-contest.org/redirect.php?banner_id=100&url=/ > <http://moblin-contest.org/redirect.php?banner_id=100&url=/>> > > _______________________________________________ > > Meshlab-devel mailing list > > Mes...@li... > <mailto:Mes...@li...> > > <mailto:Mes...@li... > <mailto:Mes...@li...>> > > https://lists.sourceforge.net/lists/listinfo/meshlab-devel > > > > > > > ------------------------------------------------------------------------ > > > > > ------------------------------------------------------------------------- > > This SF.Net email is sponsored by the Moblin Your Move > Developer's challenge > > Build the coolest Linux based applications with Moblin SDK & win > great prizes > > Grand prize is a trip for two to an Open Source event anywhere > in the world > > http://moblin-contest.org/redirect.php?banner_id=100&url=/ > <http://moblin-contest.org/redirect.php?banner_id=100&url=/> > > > ------------------------------------------------------------------------ > > > > _______________________________________________ > > Meshlab-devel mailing list > > Mes...@li... > <mailto:Mes...@li...> > > https://lists.sourceforge.net/lists/listinfo/meshlab-devel > > > > > ------------------------------------------------------------------------- > This SF.Net email is sponsored by the Moblin Your Move Developer's > challenge > Build the coolest Linux based applications with Moblin SDK & win > great prizes > Grand prize is a trip for two to an Open Source event anywhere in > the world > http://moblin-contest.org/redirect.php?banner_id=100&url=/ > <http://moblin-contest.org/redirect.php?banner_id=100&url=/> > _______________________________________________ > Meshlab-devel mailing list > Mes...@li... > <mailto:Mes...@li...> > https://lists.sourceforge.net/lists/listinfo/meshlab-devel > > > ------------------------------------------------------------------------ > > ------------------------------------------------------------------------- > This SF.Net email is sponsored by the Moblin Your Move Developer's challenge > Build the coolest Linux based applications with Moblin SDK & win great prizes > Grand prize is a trip for two to an Open Source event anywhere in the world > http://moblin-contest.org/redirect.php?banner_id=100&url=/ > > ------------------------------------------------------------------------ > > _______________________________________________ > Meshlab-devel mailing list > Mes...@li... > https://lists.sourceforge.net/lists/listinfo/meshlab-devel > -- Paolo Cignoni -- Senior Researcher Visual Computing Laboratory - ISTI - CNR http://vcg.isti.cnr.it/~cignoni ISTI - CNR Via Moruzzi 1, 56124 Pisa ITALY |
|
From: skunkwerk <sku...@gm...> - 2008-08-02 23:09:18
|
Is anyone familiar with any functions for triangulation or tesselation? because that's the way i'm thinking of splitting triangles... /vcglib/wrap/gl has a file named glu_tesselator.h i'm not sure what the "outlines" to the tesselate function are supposed to be. or which indices to pass to it - those of the model's vertices (m.vert)? the use on line 178 of import_dae.h seems to indicate that the indices are just a local, empty vector of ints and hte outlines are just the vertices - is this correct? there's also a triangulate function in hole.h, but it doesn't seem to do much any suggestions? imran On Fri, Aug 1, 2008 at 9:27 AM, skunkwerk <sku...@gm...> wrote: > thanks Fabio, i'll try it out. I was also looking for where in the code it > splits triangles that intersect the plane you're dividing with - I remember > Paolo said something about how the refine.h file can clip triangles; do you > know of any code that uses this? > > is Paolo on vacation or something? > > imran > > > On Thu, Jul 31, 2008 at 1:47 PM, fabio ganovelli < > fab...@gm...> wrote: > >> Hi, >> I think you overseen the project metro? (vcglib/apps/metro) >> In the file sampling.h of that project you can read the declaration: >> >> typedef Octree<FaceType, typename MetroMesh::ScalarType > MetroMeshOctree; >> >> and some lines later the definition: >> >> MetroMeshOctree oS2; >> >> and finally the population: >> >> if(Flags & SamplingFlags::USE_OCTREE) >> oS2.Set(S2.face.begin(),S2.face.end()); >> >> which (just like the other SI structure) is done with the function Set >> which takes >> the iterators to the first and the last element to store. >> >> However, you can also find the implementation in the file >> vcglib/space/octree.h >> (line 231) >> >> /*! >> * Populate the octree >> */ >> template < class OBJECT_ITERATOR > >> void Set(const OBJECT_ITERATOR & bObj, const OBJECT_ITERATOR & >> eObj /*, vcg::CallBackPos *callback=NULL*/) >> >> with an English comment :) >> >> cheers, >> Fabio >> >> >> >> > thanks Fabio, >> > i looked at the files you mentioned... an octree would work fine >> > for my needs, if i could figure out how to use the code in meshlab - >> > is there a function somewhere to actually populate an octree from a >> > given model? the initialize function in octree_template.h doesn't >> > seem to do much, and the comments are in Italian! >> > >> > imran >> > >> > On Thu, Jul 31, 2008 at 2:41 AM, fabio ganovelli >> > <fab...@gm... <mailto:fab...@gm...>> wrote: >> > >> > The files in /vcglib/vcg/space/index implement a few spatial >> indexing >> > data structures >> > and algorithms (regular grid, octree, spatial hashing..) with a >> common >> > interface. >> > This means that you have function to get the closest k elements in >> the >> > spatial index >> > without caring if it is a grid or an octree. See >> > vcg/complex/trimesh/closest.h to get the picture of how it is >> > done. Even >> > better, look at the project metro in vcglib/apps/metro which >> computes >> > "point to mesh" distance with (optionally) all the spatial >> > indexing data >> > structures. >> > >> > At the present the spatial indexing ds include vicinity tests and >> > (ordered) ray intersection test, but no z sorting and no generic >> BSP. >> > >> > cheers, >> > Fabio >> > >> > >> > > I'm looking to create a binary space partitioning (BSP) tree in >> > > meshlab, to use when rendering on another platform for correct >> > > z-sorting. I noticed two files already present: octree.h in >> > > /vcglib/vcg/space/index, and aabb_binary_tree.h in >> > > /vcglib/vcg/space/index (written by m_di_benedetto >> > > <http://sourceforge.net/users/m_di_benedetto/>)? Does anyone >> know >> > > what these files do, and if they could be of any use for what i'm >> > > trying to do? >> > > >> > > thanks >> > > imran >> > > >> > >> ------------------------------------------------------------------------ >> > > >> > > >> > >> ------------------------------------------------------------------------- >> > > This SF.Net email is sponsored by the Moblin Your Move >> > Developer's challenge >> > > Build the coolest Linux based applications with Moblin SDK & win >> > great prizes >> > > Grand prize is a trip for two to an Open Source event anywhere >> > in the world >> > > http://moblin-contest.org/redirect.php?banner_id=100&url=/ >> > <http://moblin-contest.org/redirect.php?banner_id=100&url=/> >> > > >> > >> ------------------------------------------------------------------------ >> > > >> > > _______________________________________________ >> > > Meshlab-devel mailing list >> > > Mes...@li... >> > <mailto:Mes...@li...> >> > > https://lists.sourceforge.net/lists/listinfo/meshlab-devel >> > > >> > >> > >> > >> ------------------------------------------------------------------------- >> > This SF.Net email is sponsored by the Moblin Your Move Developer's >> > challenge >> > Build the coolest Linux based applications with Moblin SDK & win >> > great prizes >> > Grand prize is a trip for two to an Open Source event anywhere in >> > the world >> > http://moblin-contest.org/redirect.php?banner_id=100&url=/ >> > <http://moblin-contest.org/redirect.php?banner_id=100&url=/> >> > _______________________________________________ >> > Meshlab-devel mailing list >> > Mes...@li... >> > <mailto:Mes...@li...> >> > https://lists.sourceforge.net/lists/listinfo/meshlab-devel >> > >> > >> > ------------------------------------------------------------------------ >> > >> > >> ------------------------------------------------------------------------- >> > This SF.Net email is sponsored by the Moblin Your Move Developer's >> challenge >> > Build the coolest Linux based applications with Moblin SDK & win great >> prizes >> > Grand prize is a trip for two to an Open Source event anywhere in the >> world >> > http://moblin-contest.org/redirect.php?banner_id=100&url=/ >> > ------------------------------------------------------------------------ >> > >> > _______________________________________________ >> > Meshlab-devel mailing list >> > Mes...@li... >> > https://lists.sourceforge.net/lists/listinfo/meshlab-devel >> > >> >> >> ------------------------------------------------------------------------- >> This SF.Net email is sponsored by the Moblin Your Move Developer's >> challenge >> Build the coolest Linux based applications with Moblin SDK & win great >> prizes >> Grand prize is a trip for two to an Open Source event anywhere in the >> world >> http://moblin-contest.org/redirect.php?banner_id=100&url=/ >> _______________________________________________ >> Meshlab-devel mailing list >> Mes...@li... >> https://lists.sourceforge.net/lists/listinfo/meshlab-devel >> > > |
|
From: skunkwerk <sku...@gm...> - 2008-08-01 18:50:00
|
i'm getting this error: meshlab: ../../../vcglib/vcg/complex/trimesh/update/flag.h:339: static void vcg::tri::UpdateFlags<UpdateMeshType>::FaceBorderFromNone(UpdateMeshType&) [with UpdateMeshType = CMeshO]: Assertion `p==e.end()' failed. from a call to GLA()->mm()->updateDataMask(req); the thing is, i haven't changed the mesh at all - I've just accessed its faces and vertices. so i'm not sure why this is happening.. any ideas? thanks, imran |
|
From: skunkwerk <sku...@gm...> - 2008-08-01 16:27:42
|
thanks Fabio, i'll try it out. I was also looking for where in the code it splits triangles that intersect the plane you're dividing with - I remember Paolo said something about how the refine.h file can clip triangles; do you know of any code that uses this? is Paolo on vacation or something? imran On Thu, Jul 31, 2008 at 1:47 PM, fabio ganovelli <fab...@gm...>wrote: > Hi, > I think you overseen the project metro? (vcglib/apps/metro) > In the file sampling.h of that project you can read the declaration: > > typedef Octree<FaceType, typename MetroMesh::ScalarType > MetroMeshOctree; > > and some lines later the definition: > > MetroMeshOctree oS2; > > and finally the population: > > if(Flags & SamplingFlags::USE_OCTREE) > oS2.Set(S2.face.begin(),S2.face.end()); > > which (just like the other SI structure) is done with the function Set > which takes > the iterators to the first and the last element to store. > > However, you can also find the implementation in the file > vcglib/space/octree.h > (line 231) > > /*! > * Populate the octree > */ > template < class OBJECT_ITERATOR > > void Set(const OBJECT_ITERATOR & bObj, const OBJECT_ITERATOR & > eObj /*, vcg::CallBackPos *callback=NULL*/) > > with an English comment :) > > cheers, > Fabio > > > > > thanks Fabio, > > i looked at the files you mentioned... an octree would work fine > > for my needs, if i could figure out how to use the code in meshlab - > > is there a function somewhere to actually populate an octree from a > > given model? the initialize function in octree_template.h doesn't > > seem to do much, and the comments are in Italian! > > > > imran > > > > On Thu, Jul 31, 2008 at 2:41 AM, fabio ganovelli > > <fab...@gm... <mailto:fab...@gm...>> wrote: > > > > The files in /vcglib/vcg/space/index implement a few spatial indexing > > data structures > > and algorithms (regular grid, octree, spatial hashing..) with a > common > > interface. > > This means that you have function to get the closest k elements in > the > > spatial index > > without caring if it is a grid or an octree. See > > vcg/complex/trimesh/closest.h to get the picture of how it is > > done. Even > > better, look at the project metro in vcglib/apps/metro which computes > > "point to mesh" distance with (optionally) all the spatial > > indexing data > > structures. > > > > At the present the spatial indexing ds include vicinity tests and > > (ordered) ray intersection test, but no z sorting and no generic BSP. > > > > cheers, > > Fabio > > > > > > > I'm looking to create a binary space partitioning (BSP) tree in > > > meshlab, to use when rendering on another platform for correct > > > z-sorting. I noticed two files already present: octree.h in > > > /vcglib/vcg/space/index, and aabb_binary_tree.h in > > > /vcglib/vcg/space/index (written by m_di_benedetto > > > <http://sourceforge.net/users/m_di_benedetto/>)? Does anyone know > > > what these files do, and if they could be of any use for what i'm > > > trying to do? > > > > > > thanks > > > imran > > > > > > ------------------------------------------------------------------------ > > > > > > > > > ------------------------------------------------------------------------- > > > This SF.Net email is sponsored by the Moblin Your Move > > Developer's challenge > > > Build the coolest Linux based applications with Moblin SDK & win > > great prizes > > > Grand prize is a trip for two to an Open Source event anywhere > > in the world > > > http://moblin-contest.org/redirect.php?banner_id=100&url=/ > > <http://moblin-contest.org/redirect.php?banner_id=100&url=/> > > > > > > ------------------------------------------------------------------------ > > > > > > _______________________________________________ > > > Meshlab-devel mailing list > > > Mes...@li... > > <mailto:Mes...@li...> > > > https://lists.sourceforge.net/lists/listinfo/meshlab-devel > > > > > > > > > > ------------------------------------------------------------------------- > > This SF.Net email is sponsored by the Moblin Your Move Developer's > > challenge > > Build the coolest Linux based applications with Moblin SDK & win > > great prizes > > Grand prize is a trip for two to an Open Source event anywhere in > > the world > > http://moblin-contest.org/redirect.php?banner_id=100&url=/ > > <http://moblin-contest.org/redirect.php?banner_id=100&url=/> > > _______________________________________________ > > Meshlab-devel mailing list > > Mes...@li... > > <mailto:Mes...@li...> > > https://lists.sourceforge.net/lists/listinfo/meshlab-devel > > > > > > ------------------------------------------------------------------------ > > > > ------------------------------------------------------------------------- > > This SF.Net email is sponsored by the Moblin Your Move Developer's > challenge > > Build the coolest Linux based applications with Moblin SDK & win great > prizes > > Grand prize is a trip for two to an Open Source event anywhere in the > world > > http://moblin-contest.org/redirect.php?banner_id=100&url=/ > > ------------------------------------------------------------------------ > > > > _______________________________________________ > > Meshlab-devel mailing list > > Mes...@li... > > https://lists.sourceforge.net/lists/listinfo/meshlab-devel > > > > > ------------------------------------------------------------------------- > This SF.Net email is sponsored by the Moblin Your Move Developer's > challenge > Build the coolest Linux based applications with Moblin SDK & win great > prizes > Grand prize is a trip for two to an Open Source event anywhere in the world > http://moblin-contest.org/redirect.php?banner_id=100&url=/ > _______________________________________________ > Meshlab-devel mailing list > Mes...@li... > https://lists.sourceforge.net/lists/listinfo/meshlab-devel > |
|
From: fabio g. <fab...@gm...> - 2008-07-31 20:47:24
|
Hi,
I think you overseen the project metro? (vcglib/apps/metro)
In the file sampling.h of that project you can read the declaration:
typedef Octree<FaceType, typename MetroMesh::ScalarType > MetroMeshOctree;
and some lines later the definition:
MetroMeshOctree oS2;
and finally the population:
if(Flags & SamplingFlags::USE_OCTREE)
oS2.Set(S2.face.begin(),S2.face.end());
which (just like the other SI structure) is done with the function Set
which takes
the iterators to the first and the last element to store.
However, you can also find the implementation in the file
vcglib/space/octree.h
(line 231)
/*!
* Populate the octree
*/
template < class OBJECT_ITERATOR >
void Set(const OBJECT_ITERATOR & bObj, const OBJECT_ITERATOR &
eObj /*, vcg::CallBackPos *callback=NULL*/)
with an English comment :)
cheers,
Fabio
> thanks Fabio,
> i looked at the files you mentioned... an octree would work fine
> for my needs, if i could figure out how to use the code in meshlab -
> is there a function somewhere to actually populate an octree from a
> given model? the initialize function in octree_template.h doesn't
> seem to do much, and the comments are in Italian!
>
> imran
>
> On Thu, Jul 31, 2008 at 2:41 AM, fabio ganovelli
> <fab...@gm... <mailto:fab...@gm...>> wrote:
>
> The files in /vcglib/vcg/space/index implement a few spatial indexing
> data structures
> and algorithms (regular grid, octree, spatial hashing..) with a common
> interface.
> This means that you have function to get the closest k elements in the
> spatial index
> without caring if it is a grid or an octree. See
> vcg/complex/trimesh/closest.h to get the picture of how it is
> done. Even
> better, look at the project metro in vcglib/apps/metro which computes
> "point to mesh" distance with (optionally) all the spatial
> indexing data
> structures.
>
> At the present the spatial indexing ds include vicinity tests and
> (ordered) ray intersection test, but no z sorting and no generic BSP.
>
> cheers,
> Fabio
>
>
> > I'm looking to create a binary space partitioning (BSP) tree in
> > meshlab, to use when rendering on another platform for correct
> > z-sorting. I noticed two files already present: octree.h in
> > /vcglib/vcg/space/index, and aabb_binary_tree.h in
> > /vcglib/vcg/space/index (written by m_di_benedetto
> > <http://sourceforge.net/users/m_di_benedetto/>)? Does anyone know
> > what these files do, and if they could be of any use for what i'm
> > trying to do?
> >
> > thanks
> > imran
> >
> ------------------------------------------------------------------------
> >
> >
> -------------------------------------------------------------------------
> > This SF.Net email is sponsored by the Moblin Your Move
> Developer's challenge
> > Build the coolest Linux based applications with Moblin SDK & win
> great prizes
> > Grand prize is a trip for two to an Open Source event anywhere
> in the world
> > http://moblin-contest.org/redirect.php?banner_id=100&url=/
> <http://moblin-contest.org/redirect.php?banner_id=100&url=/>
> >
> ------------------------------------------------------------------------
> >
> > _______________________________________________
> > Meshlab-devel mailing list
> > Mes...@li...
> <mailto:Mes...@li...>
> > https://lists.sourceforge.net/lists/listinfo/meshlab-devel
> >
>
>
> -------------------------------------------------------------------------
> This SF.Net email is sponsored by the Moblin Your Move Developer's
> challenge
> Build the coolest Linux based applications with Moblin SDK & win
> great prizes
> Grand prize is a trip for two to an Open Source event anywhere in
> the world
> http://moblin-contest.org/redirect.php?banner_id=100&url=/
> <http://moblin-contest.org/redirect.php?banner_id=100&url=/>
> _______________________________________________
> Meshlab-devel mailing list
> Mes...@li...
> <mailto:Mes...@li...>
> https://lists.sourceforge.net/lists/listinfo/meshlab-devel
>
>
> ------------------------------------------------------------------------
>
> -------------------------------------------------------------------------
> This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
> Build the coolest Linux based applications with Moblin SDK & win great prizes
> Grand prize is a trip for two to an Open Source event anywhere in the world
> http://moblin-contest.org/redirect.php?banner_id=100&url=/
> ------------------------------------------------------------------------
>
> _______________________________________________
> Meshlab-devel mailing list
> Mes...@li...
> https://lists.sourceforge.net/lists/listinfo/meshlab-devel
>
|
|
From: skunkwerk <sku...@gm...> - 2008-07-31 18:17:45
|
thanks Fabio, i looked at the files you mentioned... an octree would work fine for my needs, if i could figure out how to use the code in meshlab - is there a function somewhere to actually populate an octree from a given model? the initialize function in octree_template.h doesn't seem to do much, and the comments are in Italian! imran On Thu, Jul 31, 2008 at 2:41 AM, fabio ganovelli <fab...@gm...>wrote: > The files in /vcglib/vcg/space/index implement a few spatial indexing > data structures > and algorithms (regular grid, octree, spatial hashing..) with a common > interface. > This means that you have function to get the closest k elements in the > spatial index > without caring if it is a grid or an octree. See > vcg/complex/trimesh/closest.h to get the picture of how it is done. Even > better, look at the project metro in vcglib/apps/metro which computes > "point to mesh" distance with (optionally) all the spatial indexing data > structures. > > At the present the spatial indexing ds include vicinity tests and > (ordered) ray intersection test, but no z sorting and no generic BSP. > > cheers, > Fabio > > > > I'm looking to create a binary space partitioning (BSP) tree in > > meshlab, to use when rendering on another platform for correct > > z-sorting. I noticed two files already present: octree.h in > > /vcglib/vcg/space/index, and aabb_binary_tree.h in > > /vcglib/vcg/space/index (written by m_di_benedetto > > <http://sourceforge.net/users/m_di_benedetto/>)? Does anyone know > > what these files do, and if they could be of any use for what i'm > > trying to do? > > > > thanks > > imran > > ------------------------------------------------------------------------ > > > > ------------------------------------------------------------------------- > > This SF.Net email is sponsored by the Moblin Your Move Developer's > challenge > > Build the coolest Linux based applications with Moblin SDK & win great > prizes > > Grand prize is a trip for two to an Open Source event anywhere in the > world > > http://moblin-contest.org/redirect.php?banner_id=100&url=/ > > ------------------------------------------------------------------------ > > > > _______________________________________________ > > Meshlab-devel mailing list > > Mes...@li... > > https://lists.sourceforge.net/lists/listinfo/meshlab-devel > > > > > ------------------------------------------------------------------------- > This SF.Net email is sponsored by the Moblin Your Move Developer's > challenge > Build the coolest Linux based applications with Moblin SDK & win great > prizes > Grand prize is a trip for two to an Open Source event anywhere in the world > http://moblin-contest.org/redirect.php?banner_id=100&url=/ > _______________________________________________ > Meshlab-devel mailing list > Mes...@li... > https://lists.sourceforge.net/lists/listinfo/meshlab-devel > |
|
From: fabio g. <fab...@gm...> - 2008-07-31 09:41:42
|
The files in /vcglib/vcg/space/index implement a few spatial indexing data structures and algorithms (regular grid, octree, spatial hashing..) with a common interface. This means that you have function to get the closest k elements in the spatial index without caring if it is a grid or an octree. See vcg/complex/trimesh/closest.h to get the picture of how it is done. Even better, look at the project metro in vcglib/apps/metro which computes "point to mesh" distance with (optionally) all the spatial indexing data structures. At the present the spatial indexing ds include vicinity tests and (ordered) ray intersection test, but no z sorting and no generic BSP. cheers, Fabio > I'm looking to create a binary space partitioning (BSP) tree in > meshlab, to use when rendering on another platform for correct > z-sorting. I noticed two files already present: octree.h in > /vcglib/vcg/space/index, and aabb_binary_tree.h in > /vcglib/vcg/space/index (written by m_di_benedetto > <http://sourceforge.net/users/m_di_benedetto/>)? Does anyone know > what these files do, and if they could be of any use for what i'm > trying to do? > > thanks > imran > ------------------------------------------------------------------------ > > ------------------------------------------------------------------------- > This SF.Net email is sponsored by the Moblin Your Move Developer's challenge > Build the coolest Linux based applications with Moblin SDK & win great prizes > Grand prize is a trip for two to an Open Source event anywhere in the world > http://moblin-contest.org/redirect.php?banner_id=100&url=/ > ------------------------------------------------------------------------ > > _______________________________________________ > Meshlab-devel mailing list > Mes...@li... > https://lists.sourceforge.net/lists/listinfo/meshlab-devel > |
|
From: skunkwerk <sku...@gm...> - 2008-07-30 19:06:44
|
I'm looking to create a binary space partitioning (BSP) tree in meshlab, to use when rendering on another platform for correct z-sorting. I noticed two files already present: octree.h in /vcglib/vcg/space/index, and aabb_binary_tree.h in /vcglib/vcg/space/index (written by m_di_benedetto<http://sourceforge.net/users/m_di_benedetto/>)? Does anyone know what these files do, and if they could be of any use for what i'm trying to do? thanks imran |
|
From: bshu <bs...@ic...> - 2008-07-30 11:20:25
|
Dear Daniele you are right, I changed to the standard theme and it works, thanks a lot ! 2008-07-30 shu bo 发件人： Daniele Vacca 发送时间： 2008-07-30 19:10:47 收件人： mes...@li... 抄送： 主题： Re: [Meshlab-devel] my meshlab 1.1.1 when opens a model ,the model file window is quite small and can not be enla Hey bshu, I experienced the same problem few months ago... It seemed that everything was caused by my free XP theme that I've been using. I think the problem is neither Qt nor Meshlab but the patched UXTheme.dll file installed by your desktop enhancer. To put everything working, just switch to a standard but less eye candy win XP theme.. Regards /Daniele bshu wrote: Dear all: I have installed meshlab v1.1.1 on my win32 machine, os is windows xp, video card is GEFORCE 8 serires. However, when I open a 3D model using meshlab, the mesh is loaded but the window's height is quite small and can not be changed. So the mesh can not be viewed. It seems that the qt does not work well on my machine. Has anyone met the similar porblems before? could anyone help me? attached is the problem screenshot thanks! shu bo ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ _______________________________________________ Meshlab-devel mailing list Mes...@li... https://lists.sourceforge.net/lists/listinfo/meshlab-devel Windows Live Messenger Fatti rionoscere! Crea e scambia i biglietti da visita di Messenger! __________ NOD32 3308 (20080729) Information __________ This message was checked by NOD32 antivirus system. http://www.nod32cn.com |
|
From: Daniele V. <dal...@ho...> - 2008-07-30 10:51:28
|
Hey bshu, I experienced the same problem few months ago... It seemed that everything was caused by my free XP theme that I've been using. I think the problem is neither Qt nor Meshlab but the patched UXTheme.dll file installed by your desktop enhancer. To put everything working, just switch to a standard but less eye candy win XP theme.. Regards /Daniele bshu wrote: Dear all: I have installed meshlab v1.1.1 on my win32 machine, os is windows xp, video card is GEFORCE 8 serires. However, when I open a 3D model using meshlab, the mesh is loaded but the window's height is quite small and can not be changed. So the mesh can not be viewed. It seems that the qt does not work well on my machine. Has anyone met the similar porblems before? could anyone help me? attached is the problem screenshot thanks! shu bo ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ _______________________________________________ Meshlab-devel mailing list Mes...@li... https://lists.sourceforge.net/lists/listinfo/meshlab-devel _________________________________________________________________ Tutta la posta in un solo posto… scarica Windows Live Mail!!! http://get.live.com/wlmail/overview |
|
From: skunkwerk <sku...@gm...> - 2008-07-29 16:58:27
|
welcome! what's your background? what would you like to help with? imran On Tue, Jul 29, 2008 at 9:44 AM, Greg Fokum <gre...@ya...> wrote: > New member , would like to join this group > > Cheers > > GT > > > ------------------------------------------------------------------------- > This SF.Net email is sponsored by the Moblin Your Move Developer's > challenge > Build the coolest Linux based applications with Moblin SDK & win great > prizes > Grand prize is a trip for two to an Open Source event anywhere in the world > http://moblin-contest.org/redirect.php?banner_id=100&url=/ > _______________________________________________ > Meshlab-devel mailing list > Mes...@li... > https://lists.sourceforge.net/lists/listinfo/meshlab-devel > > |
|
From: Greg F. <gre...@ya...> - 2008-07-29 16:44:13
|
New member , would like to join this group
Cheers
GT
|
