Sauroposeidon in 3D

April 18, 2014

Sauroposeidon meet Sauroposeidon

I was in Oklahoma and Texas last week, seeing Sauroposeidon, Paluxysaurus, Astrophocaudia, and Alamosaurus, at the Sam Noble Oklahoma Museum of Natural History, the Fort Worth Museum of Science and History, the Shuler Museum of Paleontology at SMU, and the Perot Museum of Nature and Science, respectively. I have a ton of interesting things from that trip that I could blog about, but unfortunately I have no time. Ten days from now, I’m off to Colorado and Utah for the Mid-Mesozoic conference and field trip, and between now and then I need to finish up my bits on three collaborative papers, get my summer anatomy lectures posted for internal peer review here at WesternU, and–oh yeah–actually write my conference talk. Fun times.

BUT after being subjected to the horror of the Yale Brontosaurus skull, I figured you all deserved a little awesome.

Photographing Sauroposeidon 2014-04-07

So here’s me getting one of 351 photos of the most posterior and largest of the Sauroposeidon jackets (this is not the awesome, BTW, just a stop along the way). This jacket holds what I once inferred to be the back half of C7 and all of C8. Now that Sauroposeidon may be a somphospondyl rather than a brachiosaur, who knows what verts these are–basal somphospondyls have up to 17 cervicals to brachiosaurids’ probable 13 (for a hypothetical view of an even-longer-necked Sauroposeidon, see this probably-prophetic post by Mike). The vertically-mounted skeleton in the background is Cotylorhynchus. Cotylorhynchus got a lot bigger than that–up to maybe 6 meters long and 2 or 3 tons–and was probably the largest land animal that had ever existed back in the Early Permian. Photo by OU grad student Andrew Thomas, whom you’ll be hearing about more here in the future.

I couldn’t crank the model myself on the road, thanks to the pathetic lack of processing power in my 6-year-old laptop (which will be replaced RSN). Andy Farke volunteered to do the photogrammetricizing with Agisoft Photoscan, if only I’d DropBox him the pictures. Here’s a screenshot from MeshLab showing the result:

Sauroposeidon lateral PLY 10 - 6 and 9 blended

And my best taken-from-overhead quasi-lateral photograph:

Sauroposeidon C8 jacket lateral photo 2014-04-07

If you’re curious, the meter stick at the top is actually one meter long, it just has the English measurement side showing. The giant caliper at the bottom is also marked off in inches, and it is open to 36.0 inches (it didn’t go to 1 meter, or I would have used that). You can tell that there is some perspective distortion involved here since 36 inches on the caliper is 1380 pixels, whereas the 39.4-inch meter stick is only 1341 pixels. Man, I hate scale bars. But they make good calibration targets.

Incidentally, after playing around with the model in orthographic mode in MeshLab, the distortions in the photos of the vertebrae themselves just scream at me. Finally, finally, I can escape the tyranny of perspective. Compare the ends of the big wooden beam at the top of the jacket to get a feel for how much the two views differ.

Working on Sauroposeidon again after all this time made me seriously nostalgic. I love that beast. I don’t think I’m exaggerating when I say that those vertebrae are the most gorgeous physical objects in the universe. Also, an appropriately huge thank-you to preparator Kyle Davies (of apatosaur-sculpting fame), collections manager Jen Larsen, and Andrew Thomas again for help with wrassling those verts around, and for sharing their thoughts and advice. Thanks also to curators Rich Cifelli and Nick Czaplewski for their hospitality and for the go-ahead to undertake this work, and to Andy Farke for generating the model.

I’ll have a lot more to say about this stuff in the future. I didn’t go to all this work just for giggles. For a long time I’ve had a hankering to do a paper on the detailed anatomy of Sauroposeidon, based on all of the things that I’ve noticed in the last decade that didn’t make it into any of the early papers. And now there’s the proposed synonymy of Paluxysaurus with Sauroposeidon. And “Angloposeidon” needs some attention–Darren and I have been thinking about writing “Angloposeidon II” for years now. And…well, plenty more.

So, loads more to come, but not for the next few weeks. Eventually I’ll be publishing all of this–the photos, the 3D models, the whole works. Stay tuned.

UPDATE a few days later

Man, I am frazzled, because I forgot to include the moral of the story: if I can do this, you can do this. There are good, free photogrammetry programs out there–Peter Falkingham published a  whole paper on free photogrammetry in 2012, and posted a guide to an even better program, VisualSFM, on Even Agisoft Photoscan is not prohibitively expensive–under $200 for an educational license. MeshLab is free and has hordes of good free tutorials. For the photography itself, you basically just build a virtual dome of photos around an object. If you need more instructions than that, Heinrich has written a whole series of tutorials. It doesn’t take a fancy camera–I used a point-and-shoot for the Sauroposeidon work shown here (a Canon S100 operating at 6 megapixels, if anyone is curious). What are you waiting for?


Illustration talk slide 47

Illustration talk slide 48

Illustration talk slide 49

Illustration talk slide 50

That last one really hurts. Here’s the original image, which should have gone in the paper with the interpretive trace next to it rather than on top of it:

Sauroposeidon C6-C7 scout

The rest of the series.

Papers referenced in these slides:

Illustration talk slide 44

Illustration talk slide 45

Illustration talk slide 46

On that last slide, I also talked about two further elaborations: figures that take up the entire page, with the caption on a separate (usually facing) page, and side title figures, which are wider than tall and get turned on their sides to better use the space on the page.

Also, if I was doing this over I’d amend the statement on the last slide with, “but it doesn’t hurt you at all to be cognizant of these things, partly because they’re easy, and partly because your paper may end up at an outlet you didn’t anticipate when you wrote it.”

And I just noticed that the first slide in this group has the word ‘without’ duplicated. Jeez, what a maroon. I’ll try to remember to fix that before I post the whole slide set at the end of this exercise.

A final point: because I am picking illustrations from my whole career to illustrate these various points, almost all fail in some obvious way. The photos from the second slide should be in color, for example. When I actually gave this talk, I passed out reprints of several of my papers and said, “I am certain that every single figure I have ever made could be improved. So as you look through these papers, be thinking about how each one could be made better.”

Previous posts in this series.


Illustration talk slide 39

Illustration talk slide 40

Illustration talk slide 41

Illustration talk slide 42

Illustration talk slide 43

The Sauroposeidon stuff is cribbed from this post. For the pros and cons of scale bars in figures, see the comment thread after this post. MYDD is, of course, a thing now.

Previous posts in this series.


Wedel, M.J., and Taylor, M.P. 2013. Neural spine bifurcation in sauropod dinosaurs of the Morrison Formation: ontogenetic and phylogenetic implications. Palarch’s Journal of Vertebrate Palaeontology 10(1): 1-34. ISSN 1567-2158.

Sauroposeidon and friends

February 24, 2014

Sauroposeidon and kin cervicals - DRAFTAs a break from photography posts, here are four pretty big vertebrae that swirl in the same thought-space in my head. All are shown to scale, in right lateral view. These are not the biggest sauropod cervical vertebrae–Supersaurus beats them all, and there are vertebrae of Puertasaurus, Alamosaurus, and Futalognkosaurus that rival the big Sauroposeidon vert, but those are either less well preserved or still awaiting detailed description.

Incidentally, I think BYU 12867 is a C10. The centrum proportions are about right, compared to Giraffatitan, and the neural spine looks good, too, like a geometric transformation of the big Giraffatitan C8. Also, the drawn-in prezyg outline for MIWG.7306 is a little short; the actual prezyg is a monster and would have overhung the condyle by another 10cm or so. I’m pretty sure that we had a composite photograph showing this at one point, but irritatingly none of us can find it at the moment. If it turns up, I’ll update the image.

For a long time I thought Sauroposeidon was a brachiosaurid. Now it seems to be a somphospondyl (D’Emic 2012) or possibly even a basal titanosaur (Mannion et al. 2013), even if we stick just to the holotype. But if it’s not a brachiosaurid, it’s cervical vertebrae are at least coarsely brachiosaur-y in outline.

You  may recall from Naish et al. (2004) that MIWG.7306 shares several derived characters with the holotype vertebrae of Sauroposeidon. Does that mean that Angloposeidon is a somphospondyl or titanosaur as well? I dunno–as always, we need more material–but it’s an interesting possibility.


Following on from Matt’s post about the difficulty of photographing big specimens without distortion, I thought I’d have a play with our best Sauroposeidon C8 photo, which I think is this one:


(That’s been the basis for classic SV-POW! posts such as Your neck is pathetic and Darren’s new indeterminate Wealden maniraptoran is inadequate.)

I was motivated by Andy Farke’s comment:

Another–and perhaps more important–area where surface models excel is when you can remove colors on the original specimen that wash out relevant details…I bet this is probably the case for the example vertebra of Sauroposeidon. How many fossae and foramina just don’t show up well on the photos above?

Andy was talking about completely colourless 3d surface models, in which the 3d shape allows a render to make shadows that bring out the subtle shapes. But it made me wonder whether we could get anywhere just by washing out the most prevalent colour in the photo.

I started by doing a big, fat Gaussian blur on a duplicate layer — 500 pixels in each direction — and sampling the colour in the middle, to get a rough-and-ready average. (There may be a better way — please shout if you know one.) That average colour was#7e6b2f. I used it to run Colour To Alpha on another duplicate of the original layer, so that we’d be left with only residual colours. Here’s the result:


I’m in two minds about this. It may be informative, but it sure is ugly. To compromise, I reinstated the original layer underneath this mostly-transparent one, and turned its opacity down to 75%. Here’s the result — a nice compromise:


Of course, there are endless other approaches you can take — that’s the blessing and the curse of image-editing programs like GIMP. For example, here’s what I got doing a simple Colours → Auto → White Balance:


I’m not sure that isn’t the best of the bunch, in terms of informativeness.

I also tried something else — not amazingly successfully, but I think it’s worth seeing. Since the two photos that Matt showed in the previous post were evidently taken from somewhat different angles, I thought I’d have a go at compositing them into a red-cyan anaglyph. Because the variation in camera position is mostly dorsoventral rather than anteroposterior, the vert has to be pointed upwards for the two eyes to see the two versions from different horizontal points. Here’s the best I could do:


I would say this is of some value; but it’s nowhere near as good as, for example, the anaglyph of Cervical S of the Archbishop. I could sit and look at that one all day. The problems with this one arise for three reasons.

First, I had to reduce both parts of the Sauroposeidon anaglyph to monochrome (since one was already in that form), so all colour information was lost.

Second, I had to scale the high-resolution picture to the same size as the lower-resolution one, throwing away more detail.

Finally, and most important, the two photos were not taken with the intention that they should be used to make an anaglyph. To work well, this has to be done with the images taken under the same lighting conditions, at the same distance from the specimen, from perspectives differing by about the distance between the pupils of the viewer, and with the camera-position difference being perfectly in the plane of the specimen. Needless to say, none of these conditions was met in this case, so it’s actually quite impressive that it works as well as it does.

We have a lot of options for illustrating specimens these days. Postage-stamp-sized greyscale photos really don’t cut it any more.

Here are two photos of what I infer to be C8 of OMNH 53062, the holotype of Sauroposeidon. The top one was taken by Mike during our visit to the OMNH in 2007. If you’re a regular you may recognize it from several older posts: 1, 2, 3. The bottom one was taken by Mike Callaghan, the former museum photographer at the OMNH, sometime in 1999 or 2000. I used it in Wedel et al. (2000) and Wedel and Cifelli (2005).

Sauroposeidon OMNH 53062 C8 photos compared

You’ll notice that the two photos are far from identical. In both cases, the photographers were up on ladders, as far above the vertebra as they could get, and there are still significant perspective effects. That’s just a fact of life when you’re taking photos of a vertebra that is 1.4 meters long, from anything lower than a helicopter. In Mike Taylor’s shot, the neural spine looms a little too large; in Mike Callaghan’s shot, the prezygapophysis looks a little too small, probably because it was curving off at the edge of the shot. So neither photograph is “right”; both distort the morphology of the specimen in different ways. Here’s how the two images stack up, with the outlines scaled to the same length:

Sauroposeidon OMNH 53062 C8 outlines compared

When I ran a draft of this post past Mike, he wrote (with permission to post):

I think the current draft misses an important point: the warning. We really can’t trust photos, however carefully taken, and however beautifully composited into TNFs*. You’re welcome to quote me as having said I’d have assumed the two C8s were different vertebrae. For that matter, I bet I could have worked up several taxonomically significant characters to distinguish them. Yikes.

* TNF = Taylor Normal Form, i.e., making multi-view photos like the ones here and here.

So the moral is, photos of big specimens almost always involve some distortion. This is clearly not ideal. But I have a plan for fixing it. I am hoping to get back to the OMNH this spring, and the next time I’m there, I’m going to take photos of this vertebra from a zillion angles and make a 3D model through photogrammetry. Happily, Heinrich Mallison has been producing a very helpful series of tutorials on that very topic over at dinosaurpaleo: 1, 2, 3, 4, with more on the way (I’ll update the links here later). Update: Don’t forget to check out Peter Falkingham’s (2012) paper in PE on making photogrammetric models with free software.

Armed with that model, it should be possible to produce a perspective-free lateral view image of the vertebra, to which all of the previous photos can be compared. I can’t use CT data because this vertebra has never been CTed; it’s too big to fit through a medical CT scanner, and probably too fragile to be packed up and shipped to an industrial CT machine like they used on Sue (not to mention that would require a significant chunk of money, which is probably not worth spending on a problem that can be solved in other ways).

So, photogrammetry to the rescue, or am I just deluding myself? Let me know what you think in the comments.

Finally, I should mention that the idea of superseding photographs with 3D photogrammetric models is not original. I got religion last week while I was having beers with Martin Sander and he was showing me some of the models he’s made. He said that going forward, he was going to forbid his students to illustrate their specimens only with photographs; as far as he was concerned, now that 3D models could be cheaply and easily produced by just about everyone, they should be the new standard. Inspiring stuff–now I must go do likewise.

Some previous posts on Sauroposeidon: