I’m trying to simulate paper folding (similar to origami). It seems straightforward, but there are several issues. I initially used armature bones to drive the animation; however, when the plane folds, it results in mesh clipping.
Welcome!
If you want to make a fold using bones, you will need to do some extra manipulations to avoid perfectly folding the surface exactly on itself. I can think of 2 ways I might use.
1- Make the bend using 2 bones, add an extra short bone in between to serve as the thickness of the fold.
2- Alternatively, use a corrective shape key. Basically, you first do the fold with just one bone like you have already, then create a new shape key while in the folded position. You use that shape key to manually fix and offset the paper mesh in edit mode when it’s folded and you animate that shape key to activate when the paper is folded to correct the position created by the bones. I am not sure how well this will work with multiple folds in a chain though.
Hi , thank you for a response. I did exactly the step 1. Things do get tricky when folding multiple times. I need to try method 2. Thank you.
Hmm.. 
your “paper” does have some supdivisions in the “small bone area” ? So it does have some space between the folding.. like shown by @etn249 (..and “proper weighting”?) .
In fact this is also need when using the corrective shape..
Make sure you have Preserve Volume turned off
Hello,
When people try to make folding animation they always try to make everything in one go. But if that would be me, I’ll make in several steps with different meshes that switch to give the impression that it’s one continuous animation.
If not it becomes very challenging to make the right rig and avoid all the problems you could encounter…
Nice idea.. one often does forget that the things you see when “producing some 2D image from 3D data” is simply not real..
..so one can “freeze” any mesh in any “pose” and simple use this to “makes some more” with this “other mesh”..
So thanks for reminding this.
Yeah I think it’s a common mistake to try to recreate reality and there are many reasons to that. It’s not necessarily a bad thing either but it can makes things much more complicated than they need to be.
For instance, you want to make a werewolf transformation for your short film, you might be tempted to make a character than can turn into a werewolf and then look how you want to film that.
Or you do what is generally done : you start to make the shots/previz, then based on that you cheat each of them on a case by case basis. Basically like how these effects where done with practical effects in the 80ies.
And the same principle can go for many effects : a building that collapses , a car crash, transformers stuff …
Obviously it’s also possible to do everything in one sequence/shot too, but generally it’s harder both technically and cinematographically.
Anyway, we generally forgot how much means we have to cheat things , and also being close to reality or technically challenging doesn’t necessarily makes things better… But it’s easier to mentally represent especially when you’re beginning, that’s probably why people, especially in the beginning, try to mimic reality inside blender !
Splendid example:
One “inspiration” for this “not-make-it-as-you-see” might be.. exactly those '80 werewolf: where the hair was recorded “pulled in” and then shown in reverse.