Flying Things: A Rigging Showreel

Hi there!

I’d like to share with you my latest rigging showreel which compiles some work done at Animal Logic then Weta Digital between 2009 and 2013.

http://christophercrouzet.com/blog/post/2014/09/09/Flying-Things%3A-A-Rigging-Showreel

I mainly did it for the fun of it rather than to find a job so I aimed for something less conventional by adding a bit of derision and humour into the mix.
Let me know what you think!

Kind regards,
Christopher.

PS: I hope you don’t mind cross-posting—I couldn’t find any information in this regard here and after stumbling upon this forum only a few days ago, I thought it would be the perfect place to post this showreel!

Nice to see some humor for a change! About halfway through I was like “wait… he rigged Smaug??.. Oh…”

Fancy sharing a few pro tips for rigging a bird/wing? You seem to be the right person to ask :laugh:

Haha! That was great :slight_smile: Thanks for sharing!

The joke behind Smaug is that I was sick of rigging birds so I asked to rig him once I would finish the eagles. Didn’t happen :slight_smile:

Not sure what kind of tip to share—the idea behind the rig of the wings is relatively basic even though the implementation can be a bit tricky. In short, it’s different levels of controls driving a membrane surface which in turn drives the joints chain of each feather. Using a membrane surface has a nice side effect that allows to globally control the feathers by interpolating some attributes using the U, V coordinates.

What’s sure is that birds are definitely a pain in the ass to rig and 100% physically accurate solutions for feather simulations are yet to be found.

Interesting… I guess that makes a lot of sense from a rigging standpoint.

Kind reminds me of something used by Judd Simantov in the last of us- he made some kind of game compatible muscle system which was powered by geometry surfaces around the character.

I’ve used surfaces and curves more than once to deform organic objects.


For example, the ligament starting from the shoulder and ending onto the wrist of a bird needs to remain relatively tense when the wing is folded.
This can be represented with a NURBS curve on top of which a smooth is applied depending on the folding angle or some tension coefficient. Then it’s only a matter of deforming the underlying mesh geometry.

This kind of workflow is a bit of a hack though if the result has to be 100% anatomically correct but it still works OK and would definitely find plenty of applications in cartoons to help with some problematic deformation areas.

Cool! I’ll definitely try something like that next time I rig a bird.

I’m guessing that the animation controllers will be laid out to match the main feather groups- So you’d have a main group for the primary, tertiary and secondary and then subcontrollers which effect the membrane surfaces you mentioned?

This thread is bringing me back some memories! :slight_smile:

Yes, you need to do the distinction between each membrane in the rig because they’re independent of each other anatomically-wise. When the wings fold for example, the membranes of a same wing overlap on top of each other and you have no way of doing that if you have only a single membrane where all the feathers would sit. Even when flying, you can find some references showing how sometimes the feathers of one membrane break/split from another.


Here for example, see how on his right wing (screen left) the secondaries have a different orientation from the primaries & scapulars.


Make sure that it’s easy in the rig to bend a whole group of feathers somehow, you’ll need it both in fly and to achieve a good wing closed pose.
With a membrane surface, it’s just a matter of providing a few controllers chain along the feather direction.


On a same note, you’ll need to find a way to easily roll/twist a bunch of feathers with a gradient depending on the position of the feathers—on the photo above for example, the feathers located on the outer side of the primaries are more twisted than the ones on the inner side.
Here again, and as I said before, it can be done using the U, V coordinates of a membrane surface and applying an interpolated value depending on the U (or V) location of each feather.

The drawbacks with adding so many levels of control is that it can quickly create collisions between feathers and I don’t know any magic trick for that so it’s good to add a finer level of control over each feather for “debugging” purposes. Daniele Niero contributed a while back to develop a nice system without collisions for Legend of the Guardians but I believe it wouldn’t work with twist & roll? Maybe he’ll correct me if he passes by here :slight_smile:
Here’s the link to his article: http://danieleniero.com/wp/2013/10/12/membranae-deformer/

Last note, and as emphasized by Daniele, make sure that the feathers are not deformed by the surface like a wrap would do, but that they slide instead to preserve their shape.

Maybe I’ll post a rig one day on my blog.
Anyways, I hope it helps somehow!

Thanks for the tips! I think I’m starting to understand the basics now. I’ve never really looked into feathers with much detail before, it’s actually pretty interesting.

I also see what you mean now about birds being a pain in the ass to rig. On the bright side, at least they don’t have bloody shoulders! (Please tell me they don’t have shoulders :p:)