david's really interesting pages | Just another WordPress site

skeletal reconstructions; potential of 3D in paleoart


3D does some things with great efficiency and accuracy – things like solidity of form, staying “on-model”, etc.. Other things it doesn’t do all that well, like quickly adhering to the odd whims of its human user. These characteristics have a bit to do with the technology itself, but mostly it has to do with the interfaces used to access it.
Solidity of form is of course a big advantage when it comes to paleoart… you want that profile and front view to match, you want the jaw to rotate from where it should. So, artistically, there’s a lot going for 3D.

skeletonsBut there are also some pitfalls. As I mentioned earlier, a swell-looking skeleton can allude to accuracy that it doesn’t possess. This is comparable to the uncanny valley phenomena experienced in the representation of realistic humans. In this case, the pretense of being bone invites unintended readings – it presents general details such as surface irregularity and specularity, but lack those key elements that make that bone recognizable. Thanks but no thanks.

The beauty of technical illustrations are that they concisely filter information, representing relevant details. Here we see a skull cast of a stegosaurus, an illustration of a kiwi femur and the skeletal reconstruction by Gregory Paul that I used as reference for my 3D version. The illustrations screen information down to those elements that the artist is interested in, respecting further details only as far as they form context.

So, I set out to create a skeletal reconstruction in 3D that has the graphic clarity of those in 2D done by artists such as Gregory Paul, Scott Hartmann, Archosaurian and Dinomaniac. More than anything, this requires stylization, so that the intended information can be conveyed, but nothing more. It should immediately read as an artistic reconstruction. Hopefully, it would look good and be useful in communicating specific attributes of what makes one animal different from another, illustrating evolutionary relationships as well as functional aspects of a specific design. The stegosaurus above is my first attempt at a dinosaur, and my second after the human. I already see necessary tweaks in the skull and other areas, but I’m happy with it as a first step. Feedback welcome. Here’s a close up of the head with teeth:
Like I said. Feedback very welcome, particularly if you’re involved in paleontology.

Note: posted similar content at ArtEvolved.

There are 4 Comments to this article

Traumador the Tyrannosaur says:

this is amazing!

i’m personally finding getting palaeo feedback can be quite tricky. even if you’re working specifically with one scientist on their pet project… so much to do, so little time i think you’ll find with them PHD sorts.

how did you pull of the solid black body transparency? or is this a batch render? batch render as in first you rendered the solid black dino, and then rendered the skeleton over top of this as a background plate.

very impressive effect. hopefully it gets published in Prehistoric Times (along with all the rest of us ART Evolved ppl who submitted… you, me, and Peter from what I can gather).

admin says:

Hi Craig! Yeah… feedback is scarce. Very contact-shy these paleotypes are.

You’ll laugh when I tell you how the outline is done. Ready?
Make the polys black, low diffuse etc and… flip the polys.
Here I did render out and composite seperate passes, but even if you don’t do this there’s only minor overlap in the back and I debated with myself whether it helped make the skeleton more legible (ie. it sets the back fins further back by having the outline of the fore fins cut them off). In the end I composited as its closer to the graphic vector reconstructions that served as inspiration. You should try this. The main effort is – of course – the skeleton. But it really makes the reconstruction effort hit home.

The black volume version is transparency with a fall-off range set to depth. Depending on software package very easy or impossible.

I’m crossing my fingers (and pressing my thumbs) for us all.

Nima says:

Hay David, that’s an AWESOME Steg! Great job bringing it to 3D. It’s definitely fit for PT – in fact it’s quite possibly even TOO GOOD for PT! Reminds me of Stephen Czerkas’s famous sculpture, though I can see the Greg Paul influence as well…. one question though, the shoulder articulation looks a bit different from the Greg Paul skeletal… what happened to the coracoids, and can the humerus really articulate that way?

admin says:

Thanks Nima!
The deviation from Paul’s reconstruction is due largely to two things:
1) I made the stride more dynamic, thinking – hell, Paul always poses his theropods in a full-out run. Why not stegosaurus!
2) I don’t really know what I’m doing.

LoL! I’m learning alot, and it’d be fun to polish out some poetic reasons for this or that. But truth is, my knowledge of anatomy is solid, but artistic. It was also tough taking the tiny side and top reconstruction from the book, with countless fragmentary skeleton photos, and trying to make something loyal to a true skeleton. My main interest was the stylized abstraction of the bones, and I hope to have time in January to polish these things up a bit.

coracoids – was a combo of time running out and me not having found good reference.

shoulder articulation – I don’t have a rig yet, so I model-posed it. And it looks like the left shoulder jumped back to default pose, disarticulating from the humerus

I could use some detailed critique… maybe if I render some areas where I have questions? The skull is in need of some major re-work… likely other areas.

Write a Comment