For palaeontologists and public, the future is 3D.

Indet. ornithopod caudal vertebra recorded using photogrammetry.
480,342 polygons, with and without texture.

Here’s a prediction: 3D meshes of dinosaur bones and ichnites will become the main method of interaction with specimens for many palaeontologists within a decade. While not a panacea for all the issues that access to specimens raises for workers, this trend might go some way to mitigating some of the more persistent problems associated with the process. This prediction might come across as either a little obvious or a rather hubristic depending on your current engagement with virtual specimens and either view might be correct, but here’s why I think it will happen.

As the technology for recording, creating, distributing and working with 3D models becomes ever more widely used it’s becoming apparent that the usefulness of 3D specimen data is far more than originally envisaged. The advent of 3D printing and the ubiquity of 3D content on the internet means that meshes provide robust, testable data that is can used in a variety of ways. Biomechanics, finite element analysis and morphometrics are already well-established uses for 3D data as are comparative and descriptive purposes, reconstruction and illustration. By returning data to the physical world by 3D printing, specimens can be shared without ever having to leave the collections they are accessioned to. It’s now possible to carry and entire dinosaur skeleton on an iPad, each bone viewable at any time. Physical interaction with specimens is at the core of what we do as palaeontologists and from the field to the lab we need to be able to work on actual fossils, and this will never change. However, this might not be the most pragmatic way for us to get to all the specimens we need and 3D meshes will provide a way of accessing specimens in the future. 

Many ichnologists already do this and for them this is a no-brainer. Like many fossils, ichnites are often vulnerable to destruction and degradation, from collectors and erosion to specimens that need to remain in situ and are rightly afforded special protection so everyone can enjoy them. These can now be taken back to the lab (or pub) in as 3D data for study, recorded for research and posterity, a 3D snapshot in time. We can record how these resources change over the years, refer back to them countless times, and share them with each other easily.

For open access advocates, the rising popularity of 3D data for research should be a big deal. In an ideal world all described and figured specimens would be freely available for download at any time, by anybody. The ability to do this might have consequences for all palaeontologists, and the institutions specimens are housed in. One issue that continues to vex palaeontologists is the thorny issue of access to specimens in private and public collections. Whilst the argument that having first-hand access to a specimen is always best is irrefutable, practicalities dictate that it’s not always possible. 

Vitally, free open-access to specimens for everyone (not just scientists) might have many more benefits. The commercial trade in fossils is a subject of huge concern to any palaeontologist. With the insidious tendrils of the free market feeling their way towards more and more specimens it is a real and present danger more that more data will be lost forever; the self-regulation of markets is a myth. However, if palaeontologists can gain access to private collections or collections about to be dumped on the market by institutions (as San Deigo Museum nearly did with their Sternberg collection) all is not lost. We can now record specimens held in these collections and archive the data in a way we’ve never been able to before; there’s a tremendous amount of data out there that never gets studied. Using high-quality 3D data will remove the need for direct access the fossils to a certain degree and allow specimens otherwise unavailable to science to be worked on. We all know of specimens that are languishing in collections and might disappear at any time, lost forever and never published up. By depositing 3D data in an open access repository there will be easily obtained testable data available for research. 

This democratisation of data might have an important side effect if we could print off any fossil we wanted to. Fancy a .25 scale Tyrannosaurus rex skull on your mantlepiece? No problem, a specialist fabricators could print one off for a fraction of the price of the original. The kids want an ankylosaur spike for a school project? Download and print one off at the local high street print shop (or at home when the hardware becomes more affordable). This might be of help in pulling the rug from under a commercial fossil market that routinely prices important specimens out of the price range of virtually all public institutions; when they go, so so might the data they represent.

How could all this new data be curated? Firstly, there is now a real and present need for a cohesive set of standards to ensure the future viability of digital collections; file formats have a habit of becoming obsolete as technology races forward. Secondly, a single file format needs to be established as the standard for archiving and accessioning digital specimens. My own preference would be for Alias Wavefront’s .obj file standard for several reasons: it’s read by virtually every 3D program available, is robust, it can retain texture information in a separate jpeg file and includes the UV information relating to the texture. 

Finally, an open-access online repository to hold this data, an idea that has been wafting around for a while (see this post by Peter Falkingham for example, on establishing a database of neoichnological digital specimes). This is the biggie, and the establishment of such a resource would raise some quite contentious issues. Uploading a mesh of a dinosaur footprint spotted on a beach or of a certain outcrop is one thing, but what would be the consequences of uploading the entire holotype of a dinosaur? Museum curators might get the howling fantods at the thought of their exclusive specimens being available for free online, potentially depriving them of revenue generated from visitors to their collections. After all, this data would be freely available and if someone had the urge to print an entire skeleton out and start their own museum they would be able to, without paying a penny. Would some sort of commercial licensing be necessary? Would institutions and individuals be able to veto the inclusion of specimens they hold in such a repository? Questions, questions . . .

At the end of the day, specimens held by public institutions are owned by the public and free, unfettered access to them by anyone is the ideal scenario. Of course it’s less than ideal to allow anyone in to have a gander at your one-of-a-kind ultra-delicate fossil, but by making an accurate 3D mesh of that fossil available for all then you won’t need to; just download and print one off! For scientists and the public alike, there’s great potential in this technology if only we can all agree at the outset of how we can curate and manage it.

Not a palaeontologist? Well, you could always share heritage items such as this 3D mesh of a
2.5m recumbent stone from Arbor Low stone circle in the White Peak of Derbyshire, UK.
Send it to anyone with an internet connection. You can't do that with the real thing.

Postscript: After I wrote this Peter Falkingham posted over at his blog on the same subject, albeit from a slightly different viewpoint. I highly recommend reading it if you're interested in the digitisation of palaeontological specimens: http://pfalkingham.wordpress.com/2014/08/28/not-just-pretty-pictures/

The Photogrammetrical Palaeontologist

I don't get out into the field anywhere as much as I would like to, and when I am out there I want to be sure I can record as much data as possible, as accurately as I can. In the past this has meant using tracings, drawings, field notes, measurements and photographs to record tracks*, the location of a specimen or if excavating a specimen, the site itself. All good techniques and methods no palaeontologist worth their salt is likely to abandon in the foreseeable future, but what if you could record field localities in a way that you meant you could analyse, share with colleagues and present them back at your workstation, with sub-millimetre accuracy and all the detail visible in the field? What if you could have a scaled-down, physical section of a partially excavated skeleton from your quarry to help plan the next field season? In fact, you could be sharing not just data from the field, but specimens from collections or that you're working on right now. In the past the tools for acquiring high-quality 3D data have often been expensive and not available to all workers; we can't all afford a luggable Lidar unit to record our quarries or outcrops. As the shiny digital future becomes everyday reality a number of imaging techniques are now being used to analyse 3D data from specimens and field locations on equipment we all have, such as laptops and mobile devices.

A large ornithopod footprint in the back garden of a collector on the Isle of Wight.
Not so easy to share with your colleagues across the world? You can with photogrammetry!

Photogrammetry is one of these techniques. Most palaeontologists and amateurs will already have the tools to practice photogrammetry in their field kit: a camera and a laptop. In fact, your mobile phone and an internet connection will enable you to produce reasonable quality 3D data very little time using free software, downloadable right now.

So what is photogrammetry? It's the technique of generating a 3D point cloud from a series of overlapping photographs and at it's most basic a mesh can be generated from a stereo pair, but in most cases more images are better. It is capable of sub-millimetre accuracy and can capture virtually any subject, including outcrops and objects in the round. Photogrammetry has several advantages over traditional techniques. As mentioned earlier most of us having the equipment needed as part of our regular field kit and  vitally photogrammetry is totally non-destructive and this is important when recording delicate fossils as well as tracks and traces as often a traditional technique (for example creating a mould) will cause some damage to the fossil as part of the process. The software used to generate the 3D data is free, multi-platform, open-source or relatively cheap and capable of excellent results.

Chirotherium footprint, textured 3D mesh.

One example of a photogrammetry workflow is as follows:

1. Take overlapping photographs.
2. Load into photogrammetry application.
3. Generate a point cloud (the software looks for points on the various photos and these are assigned a point in 3D virtual space).
4. Generate mesh and texture (if required).
5. Output, analyse and share.

The 3D data has one huge advantage over traditional data: it's very easy to share. You could record a specimen in the field, generate a point cloud and the a mesh, save it into any one of a variety of formats read by a variety of apps. This data could then be emailed, uploaded and shared with colleagues across the world, all from your position in the field (provided you have internet access).

Another Isle of Wight footprint, this time a theropod track which some gooner has
tried to remove with a rocksaw. I recorded it using photogrammetry and did no damage at all.
An untextured 3D mesh.

This data has a wide variety of uses. The point cloud and mesh generated from it can be used for morphological analysis, measurements, false colour and contour analysis and light sources etc can be manipulated to aid interpretation. The meshes can also be 3D printed to bring the specimen back into the physical realm; want to have a scaled 3D version of that Allosaurus skull you excavated last field season on your desk? Use your photogrammetry data!

PG data also has potential as uses when publishing research, as stills in a paper and animations and meshes supplied as part of the supplementary data of papers. 3D works particularly well for outreach too, with animations and 3D video particularly useful for encouraging engagement.

So how to start in photogrammetry? Watch this space!

*You might notice this post is ichnology-centric. For that, I make no apology at all.

Scientific illustration: Cell schematic

Generic human cell illustration. Click to embiggen.

I've recently been working on a piece that I thought might be of interest to readers of this blog as although it's not palaeontology, who can resist a bit of biology?

This is a 3D model created in entirely in Maxon's excellent Cinema4D of a cell that is gracing the front page of my day job website (www.stupond.com), and is a schematic of a generic human cell showing the basic structures and organelles commonly present in many cells. This could be animated or labelled is needed.

Next on the personal learning curve is getting to grips with Maya, one of the real heavy-hitters in the world of 3D modelling with a learning curve to match. Exciting stuff!

I hope you like the illustration.

Using photogrammetry to capture data in the round

With DigitalFossil 2012 in a few week's time it seemed appropriate to post some of the work I've been doing recently. Part of this has focussed on creating a workflow for capturing data a full 360-degrees with minimal loss of data, which leads to holes in the mesh and distortions. By way of an illustration here is an animation of an early 360 capture of the cast of a Conchoraptor sp. skull, showing the wireframe, mesh and textured model. It loops seamlessly for your viewing pleasure.