modelling origami objects

modelling origami objects

Post by Ulf Groot » Wed, 18 Jun 2003 00:16:11



Hi, dear forum,

has anyone of you an idea of how to model paper-folded objects?

My nonexistent budget prevents me from 3D-scanning them, and I also doubt
that would produce good results since you have very deep and bended but
extremely thin "abysses" between two layers of the paper or two folds.

The last few nights I tried around with the Surface Modifier (since my
current computer is al little slow for NURBS and I thought the idea behind
the Surface Modifier and NURBS is quite the same in this case), but
encountered handling problems which might be because I did not really figure
out whether I produced valid three- or four-corner spline wireframes.
Nevertheless, reproducing very sharp folds is still nearly impossible (at
least for me), and I was unable to model even a rather simple paper plane.

Carelessly-folded parts whith unintended bends or similar I hope to cover
via bump and displacement maps, but did not make tests yet.
Also, for marco close-ups, I wonder how to give the paper a certain
thickness, since face-extruding of e.g. a Surface-modified spline does not
give me ends on both sides.

Has anyone a -maybe totally different- idea of how to model such things? Any
experience? Or any other resource I should check out?

Your advice is certainly appreciated. :)
-Ulf

 
 
 

modelling origami objects

Post by daev » Wed, 18 Jun 2003 01:08:18


I guess the long way is to fold up a plane with bend modifiers. make all
your edges on the plane the same as the creases in an unfolded object, or
your folded edges will have zig-zags where they cross edges in a rectangular
mesh. you're going to have to move individual verts around at some point,
since the bends will affect the whole mesh. this is something I've always
not had time to do, but there are many ways to do it. none I've thought up
are 'easy', my guess is you're going to spend a few hours mousing.......

good luck, and post a pic over on the binaries group if you get it to work
:)

daev


Quote:> Hi, dear forum,

> has anyone of you an idea of how to model paper-folded objects?

> My nonexistent budget prevents me from 3D-scanning them, and I also doubt
> that would produce good results since you have very deep and bended but
> extremely thin "abysses" between two layers of the paper or two folds.

> The last few nights I tried around with the Surface Modifier (since my
> current computer is al little slow for NURBS and I thought the idea behind
> the Surface Modifier and NURBS is quite the same in this case), but
> encountered handling problems which might be because I did not really
figure
> out whether I produced valid three- or four-corner spline wireframes.
> Nevertheless, reproducing very sharp folds is still nearly impossible (at
> least for me), and I was unable to model even a rather simple paper plane.

> Carelessly-folded parts whith unintended bends or similar I hope to cover
> via bump and displacement maps, but did not make tests yet.
> Also, for marco close-ups, I wonder how to give the paper a certain
> thickness, since face-extruding of e.g. a Surface-modified spline does not
> give me ends on both sides.

> Has anyone a -maybe totally different- idea of how to model such things?
Any
> experience? Or any other resource I should check out?

> Your advice is certainly appreciated. :)
> -Ulf


 
 
 

modelling origami objects

Post by napp » Wed, 18 Jun 2003 01:13:19


just fold a mesh any way you want it and use the solidify mod on it..

Quote:> Hi, dear forum,

> has anyone of you an idea of how to model paper-folded objects?

> My nonexistent budget prevents me from 3D-scanning them, and I also doubt
> that would produce good results since you have very deep and bended but
> extremely thin "abysses" between two layers of the paper or two folds.

> The last few nights I tried around with the Surface Modifier (since my
> current computer is al little slow for NURBS and I thought the idea behind
> the Surface Modifier and NURBS is quite the same in this case), but
> encountered handling problems which might be because I did not really
figure
> out whether I produced valid three- or four-corner spline wireframes.
> Nevertheless, reproducing very sharp folds is still nearly impossible (at
> least for me), and I was unable to model even a rather simple paper plane.

> Carelessly-folded parts whith unintended bends or similar I hope to cover
> via bump and displacement maps, but did not make tests yet.
> Also, for marco close-ups, I wonder how to give the paper a certain
> thickness, since face-extruding of e.g. a Surface-modified spline does not
> give me ends on both sides.

> Has anyone a -maybe totally different- idea of how to model such things?
Any
> experience? Or any other resource I should check out?

> Your advice is certainly appreciated. :)
> -Ulf

 
 
 

modelling origami objects

Post by NG » Thu, 19 Jun 2003 02:12:06


I'm not sure if this would work, but could you make a plane, with a helluva
lotta segments(so it looks solid white, instead of white lines...).  Then
select the polygons that have to be bent, put an FFD-modifier on them, and
just rotate the entire FFD cage?



Quote:> Hi, dear forum,

> has anyone of you an idea of how to model paper-folded objects?

> My nonexistent budget prevents me from 3D-scanning them, and I also doubt
> that would produce good results since you have very deep and bended but
> extremely thin "abysses" between two layers of the paper or two folds.

> The last few nights I tried around with the Surface Modifier (since my
> current computer is al little slow for NURBS and I thought the idea behind
> the Surface Modifier and NURBS is quite the same in this case), but
> encountered handling problems which might be because I did not really
figure
> out whether I produced valid three- or four-corner spline wireframes.
> Nevertheless, reproducing very sharp folds is still nearly impossible (at
> least for me), and I was unable to model even a rather simple paper plane.

> Carelessly-folded parts whith unintended bends or similar I hope to cover
> via bump and displacement maps, but did not make tests yet.
> Also, for marco close-ups, I wonder how to give the paper a certain
> thickness, since face-extruding of e.g. a Surface-modified spline does not
> give me ends on both sides.

> Has anyone a -maybe totally different- idea of how to model such things?
Any
> experience? Or any other resource I should check out?

> Your advice is certainly appreciated. :)
> -Ulf

 
 
 

modelling origami objects

Post by Ulf Groot » Thu, 19 Jun 2003 17:04:29


Unfortunately, on a real complex origami model, this (or similar mentioned)
procedure does not seem promising to me, because you got a couple of parts
on such a model which involve to bend two or three bends at once while the
rest of the model deliberately deforms during this. This gives you a round
nose on something, for example, which appears kind of solid in the paper
model, since you would not expect paper to bend like that an form a volume
body.
I have not yet considered this method along with a clever
break-up-into-small-handle-friendly-elements-plan, but I'll try.
Hey, thanks anyway - at least this post and replies prove that I was not
*that* wrong on having no quick-and-compfortable modelling idea :)



> I'm not sure if this would work, but could you make a plane, with a
helluva
> lotta segments(so it looks solid white, instead of white lines...).  Then
> select the polygons that have to be bent, put an FFD-modifier on them, and
> just rotate the entire FFD cage?



> > Hi, dear forum,

> > has anyone of you an idea of how to model paper-folded objects?

> > My nonexistent budget prevents me from 3D-scanning them, and I also
doubt
> > that would produce good results since you have very deep and bended but
> > extremely thin "abysses" between two layers of the paper or two folds.

> > The last few nights I tried around with the Surface Modifier (since my
> > current computer is al little slow for NURBS and I thought the idea
behind
> > the Surface Modifier and NURBS is quite the same in this case), but
> > encountered handling problems which might be because I did not really
> figure
> > out whether I produced valid three- or four-corner spline wireframes.
> > Nevertheless, reproducing very sharp folds is still nearly impossible
(at
> > least for me), and I was unable to model even a rather simple paper
plane.

> > Carelessly-folded parts whith unintended bends or similar I hope to
cover
> > via bump and displacement maps, but did not make tests yet.
> > Also, for marco close-ups, I wonder how to give the paper a certain
> > thickness, since face-extruding of e.g. a Surface-modified spline does
not
> > give me ends on both sides.

> > Has anyone a -maybe totally different- idea of how to model such things?
> Any
> > experience? Or any other resource I should check out?

> > Your advice is certainly appreciated. :)
> > -Ulf

 
 
 

modelling origami objects

Post by jason G2 » Fri, 20 Jun 2003 03:27:43


I just tried it, and you're right, it doesn't work...  Polygons start to
stretch, which leads to other problems...

(I finally configured my Outlook account, that's why my name's different...)



> Unfortunately, on a real complex origami model, this (or similar
mentioned)
> procedure does not seem promising to me, because you got a couple of parts
> on such a model which involve to bend two or three bends at once while the
> rest of the model deliberately deforms during this. This gives you a round
> nose on something, for example, which appears kind of solid in the paper
> model, since you would not expect paper to bend like that an form a volume
> body.
> I have not yet considered this method along with a clever
> break-up-into-small-handle-friendly-elements-plan, but I'll try.
> Hey, thanks anyway - at least this post and replies prove that I was not
> *that* wrong on having no quick-and-compfortable modelling idea :)



> > I'm not sure if this would work, but could you make a plane, with a
> helluva
> > lotta segments(so it looks solid white, instead of white lines...).
Then
> > select the polygons that have to be bent, put an FFD-modifier on them,
and
> > just rotate the entire FFD cage?



> > > Hi, dear forum,

> > > has anyone of you an idea of how to model paper-folded objects?

> > > My nonexistent budget prevents me from 3D-scanning them, and I also
> doubt
> > > that would produce good results since you have very deep and bended
but
> > > extremely thin "abysses" between two layers of the paper or two folds.

> > > The last few nights I tried around with the Surface Modifier (since my
> > > current computer is al little slow for NURBS and I thought the idea
> behind
> > > the Surface Modifier and NURBS is quite the same in this case), but
> > > encountered handling problems which might be because I did not really
> > figure
> > > out whether I produced valid three- or four-corner spline wireframes.
> > > Nevertheless, reproducing very sharp folds is still nearly impossible
> (at
> > > least for me), and I was unable to model even a rather simple paper
> plane.

> > > Carelessly-folded parts whith unintended bends or similar I hope to
> cover
> > > via bump and displacement maps, but did not make tests yet.
> > > Also, for marco close-ups, I wonder how to give the paper a certain
> > > thickness, since face-extruding of e.g. a Surface-modified spline does
> not
> > > give me ends on both sides.

> > > Has anyone a -maybe totally different- idea of how to model such
things?
> > Any
> > > experience? Or any other resource I should check out?

> > > Your advice is certainly appreciated. :)
> > > -Ulf

 
 
 

modelling origami objects

Post by Darrel Hoffma » Sun, 22 Jun 2003 04:11:45


Quote:> has anyone of you an idea of how to model paper-folded objects?

> My nonexistent budget prevents me from 3D-scanning them, and I also doubt
> that would produce good results since you have very deep and bended but
> extremely thin "abysses" between two layers of the paper or two folds.

> The last few nights I tried around with the Surface Modifier (since my
> current computer is al little slow for NURBS and I thought the idea behind
> the Surface Modifier and NURBS is quite the same in this case), but
> encountered handling problems which might be because I did not really figure
> out whether I produced valid three- or four-corner spline wireframes.
> Nevertheless, reproducing very sharp folds is still nearly impossible (at
> least for me), and I was unable to model even a rather simple paper plane.

> Carelessly-folded parts whith unintended bends or similar I hope to cover
> via bump and displacement maps, but did not make tests yet.
> Also, for marco close-ups, I wonder how to give the paper a certain
> thickness, since face-extruding of e.g. a Surface-modified spline does not
> give me ends on both sides.

> Has anyone a -maybe totally different- idea of how to model such things? Any
> experience? Or any other resource I should check out?

Myself, I'd first create the origami object in real-life, paper form.  Then just keep it in front of you while you model.  No, it's
not easy if you're not really good at this sort of thing, but for me, it's no more difficult than making a lo-poly spaceship or car.
Ignore the fact that it's folded paper and just place the vertices and faces where they are on the object.  You should be able to do
this with a fairly low poly-count, just by placing 1 vertex at each corner, and no more.  It gets a little more complicated if there
are any curved elements on the shape, but most origami I've seen is pretty much straight lines.
 
 
 

modelling origami objects

Post by Arkla » Sun, 22 Jun 2003 08:24:45


correctmy if i am wrong here... but if its made by folding... isnt EVERY
thing going to be stright, at some level?


Quote:> > has anyone of you an idea of how to model paper-folded objects?

> > My nonexistent budget prevents me from 3D-scanning them, and I also
doubt
> > that would produce good results since you have very deep and bended but
> > extremely thin "abysses" between two layers of the paper or two folds.

> > The last few nights I tried around with the Surface Modifier (since my
> > current computer is al little slow for NURBS and I thought the idea
behind
> > the Surface Modifier and NURBS is quite the same in this case), but
> > encountered handling problems which might be because I did not really
figure
> > out whether I produced valid three- or four-corner spline wireframes.
> > Nevertheless, reproducing very sharp folds is still nearly impossible
(at
> > least for me), and I was unable to model even a rather simple paper
plane.

> > Carelessly-folded parts whith unintended bends or similar I hope to
cover
> > via bump and displacement maps, but did not make tests yet.
> > Also, for marco close-ups, I wonder how to give the paper a certain
> > thickness, since face-extruding of e.g. a Surface-modified spline does
not
> > give me ends on both sides.

> > Has anyone a -maybe totally different- idea of how to model such things?
Any
> > experience? Or any other resource I should check out?

> Myself, I'd first create the origami object in real-life, paper form.

Then just keep it in front of you while you model.  No, it's
Quote:> not easy if you're not really good at this sort of thing, but for me, it's

no more difficult than making a lo-poly spaceship or car.
Quote:> Ignore the fact that it's folded paper and just place the vertices and

faces where they are on the object.  You should be able to do
Quote:> this with a fairly low poly-count, just by placing 1 vertex at each

corner, and no more.  It gets a little more complicated if there
Quote:> are any curved elements on the shape, but most origami I've seen is pretty

much straight lines.

- Show quoted text -

 
 
 

modelling origami objects

Post by daev » Sun, 22 Jun 2003 13:24:22


there are many types of folds that deform the paper in 3 dimensions, they
don't lend themselves to max very well. as an example (poorly described,
sorry, but it's hard to put into words) pulling on the corners of an area
that has been folded into quarters flattens the area in question into a
square. also there are some 'maneuvers' that involve mutiple simultaneous
folds. true, the end result will be straight lines, but the transition
involves deforming the paper in 3dimensions as it flattens out.

daev


> correctmy if i am wrong here... but if its made by folding... isnt EVERY
> thing going to be stright, at some level?



> > > has anyone of you an idea of how to model paper-folded objects?

> > > My nonexistent budget prevents me from 3D-scanning them, and I also
> doubt
> > > that would produce good results since you have very deep and bended
but
> > > extremely thin "abysses" between two layers of the paper or two folds.

> > > The last few nights I tried around with the Surface Modifier (since my
> > > current computer is al little slow for NURBS and I thought the idea
> behind
> > > the Surface Modifier and NURBS is quite the same in this case), but
> > > encountered handling problems which might be because I did not really
> figure
> > > out whether I produced valid three- or four-corner spline wireframes.
> > > Nevertheless, reproducing very sharp folds is still nearly impossible
> (at
> > > least for me), and I was unable to model even a rather simple paper
> plane.

> > > Carelessly-folded parts whith unintended bends or similar I hope to
> cover
> > > via bump and displacement maps, but did not make tests yet.
> > > Also, for marco close-ups, I wonder how to give the paper a certain
> > > thickness, since face-extruding of e.g. a Surface-modified spline does
> not
> > > give me ends on both sides.

> > > Has anyone a -maybe totally different- idea of how to model such
things?
> Any
> > > experience? Or any other resource I should check out?

> > Myself, I'd first create the origami object in real-life, paper form.
> Then just keep it in front of you while you model.  No, it's
> > not easy if you're not really good at this sort of thing, but for me,
it's
> no more difficult than making a lo-poly spaceship or car.
> > Ignore the fact that it's folded paper and just place the vertices and
> faces where they are on the object.  You should be able to do
> > this with a fairly low poly-count, just by placing 1 vertex at each
> corner, and no more.  It gets a little more complicated if there
> > are any curved elements on the shape, but most origami I've seen is
pretty
> much straight lines.

 
 
 

modelling origami objects

Post by Darrel Hoffma » Tue, 24 Jun 2003 22:22:50


Quote:> correctmy if i am wrong here... but if its made by folding... isnt EVERY
> thing going to be stright, at some level?

Not necessarily.  I've seen some origami shapes which involve curved surfaces and not just flat folds.  But my point was, I
personally would NOT start with a flat square in Max if I were modelling an origami shape.  I'd just model it already folded.  Much
more straight-foreward in my opinion.
 
 
 

modelling origami objects

Post by Arkla » Wed, 25 Jun 2003 02:49:19


im agreeing here. unless of cours,e the goal is to, dear god almight,y
animate the process of it being folded. ya know, if i knew how to do
oragami, i might just try that....


Quote:> > correctmy if i am wrong here... but if its made by folding... isnt EVERY
> > thing going to be stright, at some level?

> Not necessarily.  I've seen some origami shapes which involve curved

surfaces and not just flat folds.  But my point was, I
Quote:> personally would NOT start with a flat square in Max if I were modelling

an origami shape.  I'd just model it already folded.  Much
Quote:> more straight-foreward in my opinion.

 
 
 

modelling origami objects

Post by alla » Wed, 25 Jun 2003 05:26:40


that would be a massive undertaking, I can do a few origami things and some
of the folds that you have to do are quite complicated like multiple folds
going in multiple directs, this would be next to impossible to do in max.

I may give it a try one day.


> im agreeing here. unless of cours,e the goal is to, dear god almight,y
> animate the process of it being folded. ya know, if i knew how to do
> oragami, i might just try that....



> > > correctmy if i am wrong here... but if its made by folding... isnt
EVERY
> > > thing going to be stright, at some level?

> > Not necessarily.  I've seen some origami shapes which involve curved
> surfaces and not just flat folds.  But my point was, I
> > personally would NOT start with a flat square in Max if I were modelling
> an origami shape.  I'd just model it already folded.  Much
> > more straight-foreward in my opinion.

 
 
 

modelling origami objects

Post by Arkla » Wed, 25 Jun 2003 08:55:44


"this may be next to impossible to do in max... i may try it some day."
thats probably the funniest thing i have read this month. :)


> that would be a massive undertaking, I can do a few origami things and
some
> of the folds that you have to do are quite complicated like multiple folds
> going in multiple directs, this would be next to impossible to do in max.

> I may give it a try one day.



> > im agreeing here. unless of cours,e the goal is to, dear god almight,y
> > animate the process of it being folded. ya know, if i knew how to do
> > oragami, i might just try that....



> > > > correctmy if i am wrong here... but if its made by folding... isnt
> EVERY
> > > > thing going to be stright, at some level?

> > > Not necessarily.  I've seen some origami shapes which involve curved
> > surfaces and not just flat folds.  But my point was, I
> > > personally would NOT start with a flat square in Max if I were
modelling
> > an origami shape.  I'd just model it already folded.  Much
> > > more straight-foreward in my opinion.

 
 
 

modelling origami objects

Post by alla » Wed, 25 Jun 2003 09:04:56


gotta keep goals right LOL

> "this may be next to impossible to do in max... i may try it some day."
> thats probably the funniest thing i have read this month. :)



> > that would be a massive undertaking, I can do a few origami things and
> some
> > of the folds that you have to do are quite complicated like multiple
folds
> > going in multiple directs, this would be next to impossible to do in
max.

> > I may give it a try one day.



> > > im agreeing here. unless of cours,e the goal is to, dear god almight,y
> > > animate the process of it being folded. ya know, if i knew how to do
> > > oragami, i might just try that....



> > > > > correctmy if i am wrong here... but if its made by folding... isnt
> > EVERY
> > > > > thing going to be stright, at some level?

> > > > Not necessarily.  I've seen some origami shapes which involve curved
> > > surfaces and not just flat folds.  But my point was, I
> > > > personally would NOT start with a flat square in Max if I were
> modelling
> > > an origami shape.  I'd just model it already folded.  Much
> > > > more straight-foreward in my opinion.

 
 
 

modelling origami objects

Post by Arkla » Wed, 25 Jun 2003 11:20:55


*looks at sketch pad covered in mech and robot designs* dont remind me.


> gotta keep goals right LOL


> > "this may be next to impossible to do in max... i may try it some day."
> > thats probably the funniest thing i have read this month. :)



> > > that would be a massive undertaking, I can do a few origami things and
> > some
> > > of the folds that you have to do are quite complicated like multiple
> folds
> > > going in multiple directs, this would be next to impossible to do in
> max.

> > > I may give it a try one day.



> > > > im agreeing here. unless of cours,e the goal is to, dear god
almight,y
> > > > animate the process of it being folded. ya know, if i knew how to do
> > > > oragami, i might just try that....



> > > > > > correctmy if i am wrong here... but if its made by folding...
isnt
> > > EVERY
> > > > > > thing going to be stright, at some level?

> > > > > Not necessarily.  I've seen some origami shapes which involve
curved
> > > > surfaces and not just flat folds.  But my point was, I
> > > > > personally would NOT start with a flat square in Max if I were
> > modelling
> > > > an origami shape.  I'd just model it already folded.  Much
> > > > > more straight-foreward in my opinion.

 
 
 

1. Origami

The thought occurred to me to use the art of origami as a modelling
technique for visualization.  Sounds kind of off the wall, but I certainly
know of one serious conceptual use:

%A Haw-minn Lu
%T Computational Origami: A Geometric Approach to Regular Multiprocessing
%R MS Thesis
%I MIT
%C Cambridge, MA
%D May 1988
%X Alan Huang advisor, patent holder.

So I checked our graf-bib files and found:

%A T. Agui
%A M. Takeda
%A M. Nakajima
%T Animating planar folds by computer
%J Comput. Vision, Graphics and Image Process. (USA)
%V 24
%D Nov. 1983
%P 244-254

I think I will spend a tiny bit more time researching into this.

2. Octrees without cubes

3. maya

4. Looking For Frank Lloyd Wright Origami Chair Model

5. Volumetric Sprites.

6. modelling 3d object and motion

7. USB Mass Storage Devices (sony and iomega)

8. Modelling fluid-flow/particulate deposition around 3D objects

9. Solid Object Modelling?

10. Flexible objects modelling

11. Modelling a simple object

12. object modelling tutorial