Digitising Wildlife – Great white shark 3D photogrammetry, MRes project

February 28, 2021

So I’m doing an MRes (Master of Research, if you’ll forgive the condescension) at Staffordshire University. I’ve probably mentioned that before.

The first module was very much a primer. It introduced us to the fundamentals of research with exercises related to each stage (literature review, experiment design etc).

This concluded in a Mini-Project where we each planned, justified and executed a very small, focused research project. I completed and submitted mine during January. It would’ve been sooner, had I not been suffering from mince-pie induced lethargy.

As we received our grades this week and I can breathe a sigh of relief at not bringing shame upon my family, I’m keen to finally share the details here.

I therefore introduce you to – Digitising wildlife: Quantifying the accuracy of morphometrics captured from digital reconstructions created with 3D photogrammetry.

And before you accuse me of stuffing the title with as many big words as possible just to annoy my readers…

Well, yeah – obviously.

Story continues below...

Digitising wildlife: Quantifying the accuracy of morphometrics captured from digital reconstructions created with 3D photogrammetry.

Games and sharks. Why not?

As I’ve mentioned a fair few times already – I like sharks and videogames.

It’s long been my ambition to find a way of pairing the two and not just because I’m a child.

One of the reasons for this is that I feel games are able to engage audiences in ways no other medium can. We can become immersed in truly responsive simulations that connect us to subject matter, and present consequences to our actions in ways that just aren’t possible elsewhere.

If you don’t believe me, play This War of Mine – though I strongly suggest ordering some fluoxetine to deal with the after effects. Despair aside, there is a reason that game has been added to educational reading lists in Poland.

I therefore believe games offer a unique prospect in terms of engaging people with the natural world and some of the more complex phenomena that take place within. These could include the behavior and relationship dynamics of animals such as sharks.

The second motivation behind this pursuit is a little more technical. The games industry and the expectations of its audience have driven innovation across internet connectivity, AI simulation, content creation and real-time visualisation to name just a few areas. Games stopped being niche and became an integral part of popular culture years ago. Yet, the advantages of the associated technologies in non-games applications remain largely under utilised.

That’s true at the very least where sharks are concerned.

What a huge shock.

Doom Eternal Shark

Doom Eternal’s first piece of DLC content features a hellified shark whom shares characteristics with both the great white shark and tiger shark. It’s wonderful, but of no relevance to my project. I just think he looks neat.

So what’s the plan, Stan?

Well my name isn’t Stan, for starters.

The longterm plan is to undertake some projects which build on the principles outlined above. I’m using my MRes as a platform to begin from. While the simulation aspect of gaming is likely to become the focus of my final project, I chose to based my Mini Project around the more technical elements.

The use of 3D photogrammetry, specifically.

3D photogrammetry is something else I’ve mentioned on this website a fair few times. While it’s not strictly speaking a video game technology, its application in games is where I learned of it and have done most of my investigations.

Some folk might even remember the Virtual Reality prototype I worked on with Rich Harper and Tom Vine. We created a series of tools that allowed users to investigate digitally reconstructed sharks (among other things) with ease.

The long term potential for such a prototype as a data collection tool is largely a question of one simple thing – accuracy. So that’s what I decided to focus my project on.

Summary of the research and experiment

What follows is a super simple breakdown of the steps I took and what was learned. If you’d rather jump straight into the final report, you may download it here. We were given the option of formatting to IEEE guidelines, which I figured would be good practice.

Hence why I didn’t write it in Comic Sans.

Reconstruct, collect, compare

My basic idea was to establish how accurate morphometrics collected from a digital reconstruction are, compared to those collected in the field. Afterall, there’s no sense in praising the various hypothetical benefits of such an approach if it’s flawed at this rather fundamental level.

I also wanted to identify whether a pre-existing knowledge of the process and subject matter had any effect on this accuracy. This is because it would have great implications for the potential role of citizen science.

So using photographs very kindly provided by The Dyer Island Conservation Trust, I created a digital reconstruction of a deceased great white shark. Toby Rogers and Kelly Baker at Marine Dynamics had already collected data of this lovely fish in the field. I used said data to accurately scale the reconstruction and give myself a sample set for comparison.

Reconstruction was performed with Reality Capture and then the mesh was imported into Autodesk Maya. The maya scene comprised both the shark reconstruction and a pair of locators defining the start and end of a measurement tool.

Lastly, I created a spreadsheet which included a diagram of where each morphometric would be collected from on the shark’s body. This spreadsheet, accompanied by both some instructional videos and data set, were submitted to two clusters of participants – scientists and non-scientists.

They each very kindly and swiftly followed the guides, collected data in the spreadsheet provided and sent it back to me.

Experiment and results

On average, measurements were within 3.15CM of the physically collected morphometrics. While Scientists were generally more accurate and more consistently accurate, the most accurate measurements were taken by non-scientists.

What was particularly interesting was the skew among inaccurate results. While the lowest average deviation was 0.1CM, the highest was 8.69CM, considerably further away from the mean stated above.

I wanted to know what was causing this, so I compared the Accuracy Variation among the highest to the lowest to see if any morphometrics were particularly troublesome for participants. If data collected for specific morphometrics were of a consistently low accuracy, this would help localise investigations.

In short, it transpired that the most inaccurate data collected pertained to morphometrics whose appearance on the diagram was inconsistent with that of the reconstruction.

The below image is a perfect example. I concluded that based on the evidence, participants in such cases had to rely moreso on personal interpretation than they would in instances where the diagram and reconstruction were highly consistent. As such, a pre-existing knowledge and experience of the process would’ve provided scientists with an advantage.

If that hasn’t bored you to tears, please read more

That is a massively simplified summary, but you get the idea. I did point out numerous issues that were beyond the scope of this project, but would definitely be worth considering on a future study.

I hope to come back to this subject at a later date, but I’m also¬†really trying to adopt a more focused approach to projects (you know, so I actually finish more than none of them). Because of this, I’m likely going to focus the next stage on actual game and simulation design.

That being said, on the off-chance that you are involved with shark research or photogrammetry and would be interested in collaborating to take this concept a bit further, please do give me a nudge.

Thanks again to all involved – super appreciated.

Ta!