poissonRecon nodes order on given slice #51
Comments
|
following is a dump of active nodes of the middle slice reshape to square image , I've expected to get corners in a circle shape locations (this due to sphere ptcloud input), and got following: |
|
I’m not 100% sure I follow the question, but the nodes are ordered first by z-value (from slice to slice) and then by z-curve order (within a slice).
…-- Misha
From: daviddrezner <notifications@github.com>
Sent: Monday, April 2, 2018 9:53 AM
To: mkazhdan/PoissonRecon <PoissonRecon@noreply.github.com>
Cc: Subscribed <subscribed@noreply.github.com>
Subject: [mkazhdan/PoissonRecon] poissonRecon nodes order on given slice (#51)
Hi Michael,
Can you please explain the order of the nodes for a given slice before MC algorithm.
given to the poisson sphere ptCloud, I'm dumping nodes of the middle slice just before the MC (Marching Squares), having problem to understand the spatial order of sValues.mcIndices[i - sValues.sliceData.nodeOffset] hence corners indicator data.
Thanks,
david
—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub <#51> , or mute the thread <https://github.com/notifications/unsubscribe-auth/AKXc5RJqTmws90qzrxj8_IITKJsUpjIkks5tki04gaJpZM4TDj9u> .
|
|
Thanks Misha I will be more than grateful to get your assistance with
following,
My task is to overwrite the indicator function just following the poisson
solver, with outband values (negtive values), so the final mesh will not
pass through this voxels.
I've succeeded to overwrite some entire indicator function slice with a
negative values (outband) and to split a mesh sphere.
but I've having trouble to do it for only part of the nodes on this slice,
there are following error:
"[ERROR] Failed to close loop" _setIsoSurface() function.
drill down: my indicator function overwrite done in the
_setSliceIsoCorners() function, set squareValues to negtive value just
before MarchingSquares::GetIndex(squareValues, isoValue); function
this action produced the error that I've mentioned.
Is there a way in the released code to mask specific voxels to prevent mesh
passing through?
That's the source of my previous question regarding the order of the nodes
within the slice,
is the z-curve order is only of the active nodes? what is the relation to
the voxel absolute location of that slice.
Best Regards,
David
…On Tue, Apr 3, 2018 at 5:18 AM, mkazhdan ***@***.***> wrote:
I’m not 100% sure I follow the question, but the nodes are ordered first
by z-value (from slice to slice) and then by z-curve order (within a slice).
-- Misha
From: daviddrezner ***@***.***>
Sent: Monday, April 2, 2018 9:53 AM
To: mkazhdan/PoissonRecon ***@***.***>
Cc: Subscribed ***@***.***>
Subject: [mkazhdan/PoissonRecon] poissonRecon nodes order on given slice
(#51)
Hi Michael,
Can you please explain the order of the nodes for a given slice before MC
algorithm.
given to the poisson sphere ptCloud, I'm dumping nodes of the middle slice
just before the MC (Marching Squares), having problem to understand the
spatial order of sValues.mcIndices[i - sValues.sliceData.nodeOffset] hence
corners indicator data.
Thanks,
david
—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub <
#51> , or mute the thread <
https://github.com/notifications/unsubscribe-auth/AKXc5RJqTmws90qzrxj8_
IITKJsUpjIkks5tki04gaJpZM4TDj9u> .
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
<#51 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AkPxpdmZYoMQ4_CX2k8xy15dAc6qc1Dtks5tktvagaJpZM4TDj9u>
.
|
|
Hi David,
Unfortunately, changing it there is dangerous. At the same time that the code computes the corner values, it also computes the marching cube indices for the cells / faces. You would need to make sure that the corner values and mc indices are consistent, otherwise you can get a "failed to close loop" error.
Two other ways to do this are:
1. Extract the sampling of the indicator function on a regular grid, update values there, and then run standard marching cubes.
2. Introduce additional point constraints to force the function to have prescribed values at specific locations. (I know this will be supported in the soon-to-be released version. Not sure if it's supported in the current version.)
…-- Misha
On April 3, 2018 3:39:17 AM EDT, daviddrezner ***@***.***> wrote:
Thanks Misha I will be more than grateful to get your assistance with
following,
My task is to overwrite the indicator function just following the
poisson
solver, with outband values (negtive values), so the final mesh will
not
pass through this voxels.
I've succeeded to overwrite some entire indicator function slice with a
negative values (outband) and to split a mesh sphere.
but I've having trouble to do it for only part of the nodes on this
slice,
there are following error:
"[ERROR] Failed to close loop" _setIsoSurface() function.
drill down: my indicator function overwrite done in the
_setSliceIsoCorners() function, set squareValues to negtive value
just
before MarchingSquares::GetIndex(squareValues, isoValue); function
this action produced the error that I've mentioned.
Is there a way in the released code to mask specific voxels to prevent
mesh
passing through?
That's the source of my previous question regarding the order of the
nodes
within the slice,
is the z-curve order is only of the active nodes? what is the relation
to
the voxel absolute location of that slice.
Best Regards,
David
On Tue, Apr 3, 2018 at 5:18 AM, mkazhdan ***@***.***>
wrote:
> I’m not 100% sure I follow the question, but the nodes are ordered
first
> by z-value (from slice to slice) and then by z-curve order (within a
slice).
>
>
>
> -- Misha
>
>
>
> From: daviddrezner ***@***.***>
> Sent: Monday, April 2, 2018 9:53 AM
> To: mkazhdan/PoissonRecon ***@***.***>
> Cc: Subscribed ***@***.***>
> Subject: [mkazhdan/PoissonRecon] poissonRecon nodes order on given
slice
> (#51)
>
>
>
> Hi Michael,
>
> Can you please explain the order of the nodes for a given slice
before MC
> algorithm.
> given to the poisson sphere ptCloud, I'm dumping nodes of the middle
slice
> just before the MC (Marching Squares), having problem to understand
the
> spatial order of sValues.mcIndices[i - sValues.sliceData.nodeOffset]
hence
> corners indicator data.
>
> Thanks,
> david
>
> —
> You are receiving this because you are subscribed to this thread.
> Reply to this email directly, view it on GitHub <
> #51> , or mute the
thread <
>
https://github.com/notifications/unsubscribe-auth/AKXc5RJqTmws90qzrxj8_
> IITKJsUpjIkks5tki04gaJpZM4TDj9u> .
>
> —
> You are receiving this because you authored the thread.
> Reply to this email directly, view it on GitHub
>
<#51 (comment)>,
> or mute the thread
>
<https://github.com/notifications/unsubscribe-auth/AkPxpdmZYoMQ4_CX2k8xy15dAc6qc1Dtks5tktvagaJpZM4TDj9u>
> .
>
--
You are receiving this because you commented.
Reply to this email directly or view it on GitHub:
#51 (comment)
|
|
Thanks Misha for all your help,
Great to hear regarding this capability in the coming version,
Can you please estimate when this version with the additional points
constrains would be released (days / weeks)?
Best Regards,
David
…On Tue, Apr 3, 2018 at 3:24 PM, mkazhdan ***@***.***> wrote:
Hi David,
Unfortunately, changing it there is dangerous. At the same time that the
code computes the corner values, it also computes the marching cube indices
for the cells / faces. You would need to make sure that the corner values
and mc indices are consistent, otherwise you can get a "failed to close
loop" error.
Two other ways to do this are:
1. Extract the sampling of the indicator function on a regular grid,
update values there, and then run standard marching cubes.
2. Introduce additional point constraints to force the function to have
prescribed values at specific locations. (I know this will be supported in
the soon-to-be released version. Not sure if it's supported in the current
version.)
-- Misha
On April 3, 2018 3:39:17 AM EDT, daviddrezner ***@***.***>
wrote:
>Thanks Misha I will be more than grateful to get your assistance with
>following,
>
>My task is to overwrite the indicator function just following the
>poisson
>solver, with outband values (negtive values), so the final mesh will
>not
>pass through this voxels.
>
>I've succeeded to overwrite some entire indicator function slice with a
>negative values (outband) and to split a mesh sphere.
>but I've having trouble to do it for only part of the nodes on this
>slice,
>there are following error:
>"[ERROR] Failed to close loop" _setIsoSurface() function.
>
>drill down: my indicator function overwrite done in the
>_setSliceIsoCorners() function, set squareValues to negtive value
>just
>before MarchingSquares::GetIndex(squareValues, isoValue); function
>this action produced the error that I've mentioned.
>
>Is there a way in the released code to mask specific voxels to prevent
>mesh
>passing through?
>
>That's the source of my previous question regarding the order of the
>nodes
>within the slice,
>is the z-curve order is only of the active nodes? what is the relation
>to
>the voxel absolute location of that slice.
>
>Best Regards,
>David
>
>On Tue, Apr 3, 2018 at 5:18 AM, mkazhdan ***@***.***>
>wrote:
>
>> I’m not 100% sure I follow the question, but the nodes are ordered
>first
>> by z-value (from slice to slice) and then by z-curve order (within a
>slice).
>>
>>
>>
>> -- Misha
>>
>>
>>
>> From: daviddrezner ***@***.***>
>> Sent: Monday, April 2, 2018 9:53 AM
>> To: mkazhdan/PoissonRecon ***@***.***>
>> Cc: Subscribed ***@***.***>
>> Subject: [mkazhdan/PoissonRecon] poissonRecon nodes order on given
>slice
>> (#51)
>>
>>
>>
>> Hi Michael,
>>
>> Can you please explain the order of the nodes for a given slice
>before MC
>> algorithm.
>> given to the poisson sphere ptCloud, I'm dumping nodes of the middle
>slice
>> just before the MC (Marching Squares), having problem to understand
>the
>> spatial order of sValues.mcIndices[i - sValues.sliceData.nodeOffset]
>hence
>> corners indicator data.
>>
>> Thanks,
>> david
>>
>> —
>> You are receiving this because you are subscribed to this thread.
>> Reply to this email directly, view it on GitHub <
>> #51> , or mute the
>thread <
>>
>https://github.com/notifications/unsubscribe-auth/AKXc5RJqTmws90qzrxj8_
>> IITKJsUpjIkks5tki04gaJpZM4TDj9u> .
>>
>> —
>> You are receiving this because you authored the thread.
>> Reply to this email directly, view it on GitHub
>>
><#51#
issuecomment-378106424>,
>> or mute the thread
>>
><https://github.com/notifications/unsubscribe-auth/AkPxpdmZYoMQ4_
CX2k8xy15dAc6qc1Dtks5tktvagaJpZM4TDj9u>
>> .
>>
>
>
>--
>You are receiving this because you commented.
>Reply to this email directly or view it on GitHub:
>#51 (comment)
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
<#51 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AkPxpQVD7wngZL89O1IkFim6rnqKE_bPks5tk2oagaJpZM4TDj9u>
.
|
|
Hi Misha,
Regarding the additional point which constraints the function at specific
locations, the format of this points would be as the input point cloud, but
without normals and with indicator function value constrain?
It will also force the octree to have the needed depth (up to maxdepth)?
regarding your option 1 " Extract the sampling of the indicator function on
a regular grid, update values there, and then run standard marching cubes.
" can you please share the extract function to a regular grid with me ,
or is it exist in the poissonRecon code?
I like to offer my help with this new release, I'm experienced alg eng. and
programmer.
Best Regards,
David
…On Tue, Apr 3, 2018 at 4:46 PM, david drezner ***@***.***> wrote:
Thanks Misha for all your help,
Great to hear regarding this capability in the coming version,
Can you please estimate when this version with the additional points
constrains would be released (days / weeks)?
Best Regards,
David
On Tue, Apr 3, 2018 at 3:24 PM, mkazhdan ***@***.***> wrote:
> Hi David,
>
> Unfortunately, changing it there is dangerous. At the same time that the
> code computes the corner values, it also computes the marching cube indices
> for the cells / faces. You would need to make sure that the corner values
> and mc indices are consistent, otherwise you can get a "failed to close
> loop" error.
>
> Two other ways to do this are:
> 1. Extract the sampling of the indicator function on a regular grid,
> update values there, and then run standard marching cubes.
> 2. Introduce additional point constraints to force the function to have
> prescribed values at specific locations. (I know this will be supported in
> the soon-to-be released version. Not sure if it's supported in the current
> version.)
>
> -- Misha
>
>
> On April 3, 2018 3:39:17 AM EDT, daviddrezner ***@***.***>
> wrote:
> >Thanks Misha I will be more than grateful to get your assistance with
> >following,
> >
> >My task is to overwrite the indicator function just following the
> >poisson
> >solver, with outband values (negtive values), so the final mesh will
> >not
> >pass through this voxels.
> >
> >I've succeeded to overwrite some entire indicator function slice with a
> >negative values (outband) and to split a mesh sphere.
> >but I've having trouble to do it for only part of the nodes on this
> >slice,
> >there are following error:
> >"[ERROR] Failed to close loop" _setIsoSurface() function.
> >
> >drill down: my indicator function overwrite done in the
> >_setSliceIsoCorners() function, set squareValues to negtive value
> >just
> >before MarchingSquares::GetIndex(squareValues, isoValue); function
> >this action produced the error that I've mentioned.
> >
> >Is there a way in the released code to mask specific voxels to prevent
> >mesh
> >passing through?
> >
> >That's the source of my previous question regarding the order of the
> >nodes
> >within the slice,
> >is the z-curve order is only of the active nodes? what is the relation
> >to
> >the voxel absolute location of that slice.
> >
> >Best Regards,
> >David
> >
> >On Tue, Apr 3, 2018 at 5:18 AM, mkazhdan ***@***.***>
> >wrote:
> >
> >> I’m not 100% sure I follow the question, but the nodes are ordered
> >first
> >> by z-value (from slice to slice) and then by z-curve order (within a
> >slice).
> >>
> >>
> >>
> >> -- Misha
> >>
> >>
> >>
> >> From: daviddrezner ***@***.***>
> >> Sent: Monday, April 2, 2018 9:53 AM
> >> To: mkazhdan/PoissonRecon ***@***.***>
> >> Cc: Subscribed ***@***.***>
> >> Subject: [mkazhdan/PoissonRecon] poissonRecon nodes order on given
> >slice
> >> (#51)
> >>
> >>
> >>
> >> Hi Michael,
> >>
> >> Can you please explain the order of the nodes for a given slice
> >before MC
> >> algorithm.
> >> given to the poisson sphere ptCloud, I'm dumping nodes of the middle
> >slice
> >> just before the MC (Marching Squares), having problem to understand
> >the
> >> spatial order of sValues.mcIndices[i - sValues.sliceData.nodeOffset]
> >hence
> >> corners indicator data.
> >>
> >> Thanks,
> >> david
> >>
> >> —
> >> You are receiving this because you are subscribed to this thread.
> >> Reply to this email directly, view it on GitHub <
> >> #51> , or mute the
> >thread <
> >>
> >https://github.com/notifications/unsubscribe-auth/AKXc5RJqTmws90qzrxj8_
> >> IITKJsUpjIkks5tki04gaJpZM4TDj9u> .
> >>
> >> —
> >> You are receiving this because you authored the thread.
> >> Reply to this email directly, view it on GitHub
> >>
> ><#51 (comment)
> mment-378106424>,
> >> or mute the thread
> >>
> ><https://github.com/notifications/unsubscribe-auth/
> AkPxpdmZYoMQ4_CX2k8xy15dAc6qc1Dtks5tktvagaJpZM4TDj9u>
> >> .
> >>
> >
> >
> >--
> >You are receiving this because you commented.
> >Reply to this email directly or view it on GitHub:
> >#51 (comment)
> ment-378157496
>
> —
> You are receiving this because you authored the thread.
> Reply to this email directly, view it on GitHub
> <#51 (comment)>,
> or mute the thread
> <https://github.com/notifications/unsubscribe-auth/AkPxpQVD7wngZL89O1IkFim6rnqKE_bPks5tk2oagaJpZM4TDj9u>
> .
>
|
|
The new code will support the functionality for point constraints, though getting a front-end will take a bit of work. (Points won't require normals. Points will take constraint values. And points will be added at the finest depth.)
For the regular grid, you can use the --voxel option. (If you need an implementation for surface extraction from a regular grid, you can grab the code from http://www.cs.jhu.edu/~misha/Code/IsoSurfaceExtraction/)
…-- Misha
On April 4, 2018 8:27:03 AM EDT, daviddrezner ***@***.***> wrote:
Hi Misha,
Regarding the additional point which constraints the function at
specific
locations, the format of this points would be as the input point cloud,
but
without normals and with indicator function value constrain?
It will also force the octree to have the needed depth (up to
maxdepth)?
regarding your option 1 " Extract the sampling of the indicator
function on
a regular grid, update values there, and then run standard marching
cubes.
" can you please share the extract function to a regular grid with
me ,
or is it exist in the poissonRecon code?
I like to offer my help with this new release, I'm experienced alg eng.
and
programmer.
Best Regards,
David
On Tue, Apr 3, 2018 at 4:46 PM, david drezner ***@***.***> wrote:
> Thanks Misha for all your help,
>
> Great to hear regarding this capability in the coming version,
> Can you please estimate when this version with the additional points
> constrains would be released (days / weeks)?
>
> Best Regards,
> David
>
>
>
> On Tue, Apr 3, 2018 at 3:24 PM, mkazhdan ***@***.***>
wrote:
>
>> Hi David,
>>
>> Unfortunately, changing it there is dangerous. At the same time that
the
>> code computes the corner values, it also computes the marching cube
indices
>> for the cells / faces. You would need to make sure that the corner
values
>> and mc indices are consistent, otherwise you can get a "failed to
close
>> loop" error.
>>
>> Two other ways to do this are:
>> 1. Extract the sampling of the indicator function on a regular grid,
>> update values there, and then run standard marching cubes.
>> 2. Introduce additional point constraints to force the function to
have
>> prescribed values at specific locations. (I know this will be
supported in
>> the soon-to-be released version. Not sure if it's supported in the
current
>> version.)
>>
>> -- Misha
>>
>>
>> On April 3, 2018 3:39:17 AM EDT, daviddrezner
***@***.***>
>> wrote:
>> >Thanks Misha I will be more than grateful to get your assistance
with
>> >following,
>> >
>> >My task is to overwrite the indicator function just following the
>> >poisson
>> >solver, with outband values (negtive values), so the final mesh
will
>> >not
>> >pass through this voxels.
>> >
>> >I've succeeded to overwrite some entire indicator function slice
with a
>> >negative values (outband) and to split a mesh sphere.
>> >but I've having trouble to do it for only part of the nodes on this
>> >slice,
>> >there are following error:
>> >"[ERROR] Failed to close loop" _setIsoSurface() function.
>> >
>> >drill down: my indicator function overwrite done in the
>> >_setSliceIsoCorners() function, set squareValues to negtive value
>> >just
>> >before MarchingSquares::GetIndex(squareValues, isoValue); function
>> >this action produced the error that I've mentioned.
>> >
>> >Is there a way in the released code to mask specific voxels to
prevent
>> >mesh
>> >passing through?
>> >
>> >That's the source of my previous question regarding the order of
the
>> >nodes
>> >within the slice,
>> >is the z-curve order is only of the active nodes? what is the
relation
>> >to
>> >the voxel absolute location of that slice.
>> >
>> >Best Regards,
>> >David
>> >
>> >On Tue, Apr 3, 2018 at 5:18 AM, mkazhdan ***@***.***>
>> >wrote:
>> >
>> >> I’m not 100% sure I follow the question, but the nodes are
ordered
>> >first
>> >> by z-value (from slice to slice) and then by z-curve order
(within a
>> >slice).
>> >>
>> >>
>> >>
>> >> -- Misha
>> >>
>> >>
>> >>
>> >> From: daviddrezner ***@***.***>
>> >> Sent: Monday, April 2, 2018 9:53 AM
>> >> To: mkazhdan/PoissonRecon ***@***.***>
>> >> Cc: Subscribed ***@***.***>
>> >> Subject: [mkazhdan/PoissonRecon] poissonRecon nodes order on
given
>> >slice
>> >> (#51)
>> >>
>> >>
>> >>
>> >> Hi Michael,
>> >>
>> >> Can you please explain the order of the nodes for a given slice
>> >before MC
>> >> algorithm.
>> >> given to the poisson sphere ptCloud, I'm dumping nodes of the
middle
>> >slice
>> >> just before the MC (Marching Squares), having problem to
understand
>> >the
>> >> spatial order of sValues.mcIndices[i -
sValues.sliceData.nodeOffset]
>> >hence
>> >> corners indicator data.
>> >>
>> >> Thanks,
>> >> david
>> >>
>> >> —
>> >> You are receiving this because you are subscribed to this thread.
>> >> Reply to this email directly, view it on GitHub <
>> >> #51> , or mute the
>> >thread <
>> >>
>>
>https://github.com/notifications/unsubscribe-auth/AKXc5RJqTmws90qzrxj8_
>> >> IITKJsUpjIkks5tki04gaJpZM4TDj9u> .
>> >>
>> >> —
>> >> You are receiving this because you authored the thread.
>> >> Reply to this email directly, view it on GitHub
>> >>
>> ><#51 (comment)
>> mment-378106424>,
>> >> or mute the thread
>> >>
>> ><https://github.com/notifications/unsubscribe-auth/
>> AkPxpdmZYoMQ4_CX2k8xy15dAc6qc1Dtks5tktvagaJpZM4TDj9u>
>> >> .
>> >>
>> >
>> >
>> >--
>> >You are receiving this because you commented.
>> >Reply to this email directly or view it on GitHub:
>> >#51 (comment)
>> ment-378157496
>>
>> —
>> You are receiving this because you authored the thread.
>> Reply to this email directly, view it on GitHub
>>
<#51 (comment)>,
>> or mute the thread
>>
<https://github.com/notifications/unsubscribe-auth/AkPxpQVD7wngZL89O1IkFim6rnqKE_bPks5tk2oagaJpZM4TDj9u>
>> .
>>
>
>
--
You are receiving this because you commented.
Reply to this email directly or view it on GitHub:
#51 (comment)
|
|
Thanks a lot Misha for all your help
…On Wed, Apr 4, 2018 at 5:09 PM, mkazhdan ***@***.***> wrote:
The new code will support the functionality for point constraints, though
getting a front-end will take a bit of work. (Points won't require normals.
Points will take constraint values. And points will be added at the finest
depth.)
For the regular grid, you can use the --voxel option. (If you need an
implementation for surface extraction from a regular grid, you can grab the
code from http://www.cs.jhu.edu/~misha/Code/IsoSurfaceExtraction/)
-- Misha
On April 4, 2018 8:27:03 AM EDT, daviddrezner ***@***.***>
wrote:
>Hi Misha,
>
>Regarding the additional point which constraints the function at
>specific
>locations, the format of this points would be as the input point cloud,
>but
>without normals and with indicator function value constrain?
>It will also force the octree to have the needed depth (up to
>maxdepth)?
>
>regarding your option 1 " Extract the sampling of the indicator
>function on
>a regular grid, update values there, and then run standard marching
>cubes.
>" can you please share the extract function to a regular grid with
>me ,
>or is it exist in the poissonRecon code?
>
>I like to offer my help with this new release, I'm experienced alg eng.
>and
>programmer.
>
>Best Regards,
>David
>
>On Tue, Apr 3, 2018 at 4:46 PM, david drezner ***@***.***> wrote:
>
>> Thanks Misha for all your help,
>>
>> Great to hear regarding this capability in the coming version,
>> Can you please estimate when this version with the additional points
>> constrains would be released (days / weeks)?
>>
>> Best Regards,
>> David
>>
>>
>>
>> On Tue, Apr 3, 2018 at 3:24 PM, mkazhdan ***@***.***>
>wrote:
>>
>>> Hi David,
>>>
>>> Unfortunately, changing it there is dangerous. At the same time that
>the
>>> code computes the corner values, it also computes the marching cube
>indices
>>> for the cells / faces. You would need to make sure that the corner
>values
>>> and mc indices are consistent, otherwise you can get a "failed to
>close
>>> loop" error.
>>>
>>> Two other ways to do this are:
>>> 1. Extract the sampling of the indicator function on a regular grid,
>>> update values there, and then run standard marching cubes.
>>> 2. Introduce additional point constraints to force the function to
>have
>>> prescribed values at specific locations. (I know this will be
>supported in
>>> the soon-to-be released version. Not sure if it's supported in the
>current
>>> version.)
>>>
>>> -- Misha
>>>
>>>
>>> On April 3, 2018 3:39:17 AM EDT, daviddrezner
***@***.***>
>>> wrote:
>>> >Thanks Misha I will be more than grateful to get your assistance
>with
>>> >following,
>>> >
>>> >My task is to overwrite the indicator function just following the
>>> >poisson
>>> >solver, with outband values (negtive values), so the final mesh
>will
>>> >not
>>> >pass through this voxels.
>>> >
>>> >I've succeeded to overwrite some entire indicator function slice
>with a
>>> >negative values (outband) and to split a mesh sphere.
>>> >but I've having trouble to do it for only part of the nodes on this
>>> >slice,
>>> >there are following error:
>>> >"[ERROR] Failed to close loop" _setIsoSurface() function.
>>> >
>>> >drill down: my indicator function overwrite done in the
>>> >_setSliceIsoCorners() function, set squareValues to negtive value
>>> >just
>>> >before MarchingSquares::GetIndex(squareValues, isoValue); function
>>> >this action produced the error that I've mentioned.
>>> >
>>> >Is there a way in the released code to mask specific voxels to
>prevent
>>> >mesh
>>> >passing through?
>>> >
>>> >That's the source of my previous question regarding the order of
>the
>>> >nodes
>>> >within the slice,
>>> >is the z-curve order is only of the active nodes? what is the
>relation
>>> >to
>>> >the voxel absolute location of that slice.
>>> >
>>> >Best Regards,
>>> >David
>>> >
>>> >On Tue, Apr 3, 2018 at 5:18 AM, mkazhdan ***@***.***>
>>> >wrote:
>>> >
>>> >> I’m not 100% sure I follow the question, but the nodes are
>ordered
>>> >first
>>> >> by z-value (from slice to slice) and then by z-curve order
>(within a
>>> >slice).
>>> >>
>>> >>
>>> >>
>>> >> -- Misha
>>> >>
>>> >>
>>> >>
>>> >> From: daviddrezner ***@***.***>
>>> >> Sent: Monday, April 2, 2018 9:53 AM
>>> >> To: mkazhdan/PoissonRecon ***@***.***>
>>> >> Cc: Subscribed ***@***.***>
>>> >> Subject: [mkazhdan/PoissonRecon] poissonRecon nodes order on
>given
>>> >slice
>>> >> (#51)
>>> >>
>>> >>
>>> >>
>>> >> Hi Michael,
>>> >>
>>> >> Can you please explain the order of the nodes for a given slice
>>> >before MC
>>> >> algorithm.
>>> >> given to the poisson sphere ptCloud, I'm dumping nodes of the
>middle
>>> >slice
>>> >> just before the MC (Marching Squares), having problem to
>understand
>>> >the
>>> >> spatial order of sValues.mcIndices[i -
>sValues.sliceData.nodeOffset]
>>> >hence
>>> >> corners indicator data.
>>> >>
>>> >> Thanks,
>>> >> david
>>> >>
>>> >> —
>>> >> You are receiving this because you are subscribed to this thread.
>>> >> Reply to this email directly, view it on GitHub <
>>> >> #51> , or mute the
>>> >thread <
>>> >>
>>>
>>https://github.com/notifications/unsubscribe-auth/AKXc5RJqTmws90qzrxj8_
>>> >> IITKJsUpjIkks5tki04gaJpZM4TDj9u> .
>>> >>
>>> >> —
>>> >> You are receiving this because you authored the thread.
>>> >> Reply to this email directly, view it on GitHub
>>> >>
>>> ><#51 (comment)
>>> mment-378106424>,
>>> >> or mute the thread
>>> >>
>>> ><https://github.com/notifications/unsubscribe-auth/
>>> AkPxpdmZYoMQ4_CX2k8xy15dAc6qc1Dtks5tktvagaJpZM4TDj9u>
>>> >> .
>>> >>
>>> >
>>> >
>>> >--
>>> >You are receiving this because you commented.
>>> >Reply to this email directly or view it on GitHub:
>>> >#51 (comment)
>>> ment-378157496
>>>
>>> —
>>> You are receiving this because you authored the thread.
>>> Reply to this email directly, view it on GitHub
>>>
><#51#
issuecomment-378231352>,
>>> or mute the thread
>>>
><https://github.com/notifications/unsubscribe-auth/
AkPxpQVD7wngZL89O1IkFim6rnqKE_bPks5tk2oagaJpZM4TDj9u>
>>> .
>>>
>>
>>
>
>
>--
>You are receiving this because you commented.
>Reply to this email directly or view it on GitHub:
>#51 (comment)
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
<#51 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AkPxpWzwBRGUSy4tcuVtXIb410ykYGg7ks5tlNQmgaJpZM4TDj9u>
.
|
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Hi Michael,
Can you please explain the order of the nodes for a given slice before MC algorithm.
given to the poisson sphere ptCloud, I'm dumping nodes of the middle slice just before the MC (Marching Squares), having problem to understand the spatial order of sValues.mcIndices[i - sValues.sliceData.nodeOffset] hence corners indicator data.
Thanks,
david
The text was updated successfully, but these errors were encountered: