Search Unity

  1. Megacity Metro Demo now available. Download now.
    Dismiss Notice
  2. Unity support for visionOS is now available. Learn more in our blog post.
    Dismiss Notice

Zoological data?

Discussion in 'General Discussion' started by yoonitee, Oct 4, 2018.

  1. yoonitee

    yoonitee

    Joined:
    Jun 27, 2013
    Posts:
    2,363
    I wonder if anyone knows any free data sources for zoological data? For example digitised measurements of bones and skeletons of different animals. It would seem like a useful thing to exist and useful for making animal models for games. But I dont think it does exist. Alternatives might be a database of photographs from different angles of animal skeletons or dinosaur skeletons.

    Basically Im looking for a database that i can process and get morphological data about hundreds of animals.
     
  2. lipisis

    lipisis

    Joined:
    Jan 14, 2017
    Posts:
    37
    Not exactly bones but all in all some animal size and stuff.
    http://esapubs.org/archive/ecol/E090/184/

    I found after 3 mins of google. You should search a little bit more and I'm sure you can find much more.
     
  3. neoshaman

    neoshaman

    Joined:
    Feb 11, 2011
    Posts:
    6,492
  4. yoonitee

    yoonitee

    Joined:
    Jun 27, 2013
    Posts:
    2,363
    I did lots of searching. Couldn't find much at all. I would have thought that every natural history museum would be digitising its collections and scanning them in but if they are the datas not on the internet! Cant even find a database of photos of animal skeletons.
     
  5. neoshaman

    neoshaman

    Joined:
    Feb 11, 2011
    Posts:
    6,492
    I just looked for 3d scan bones animal as key words (also image tab)
     
  6. yoonitee

    yoonitee

    Joined:
    Jun 27, 2013
    Posts:
    2,363
    Its good but the problem here is that they're mainly fragments or parts of skeletons. Im really just looking for skeletons of whole modern animals.

    I could probably find images of each animal one by one. But a catalogue would be better.
     
  7. neoshaman

    neoshaman

    Joined:
    Feb 11, 2011
    Posts:
    6,492
  8. yoonitee

    yoonitee

    Joined:
    Jun 27, 2013
    Posts:
    2,363
    Last edited: Oct 5, 2018
  9. BIGTIMEMASTER

    BIGTIMEMASTER

    Joined:
    Jun 1, 2017
    Posts:
    5,181
    https://www.artstation.com/artwork/612Z6

    There's a leopard skeleton.

    I'd be interested to find a comprehensive database, but I doubt such exist. Probably best bet is to go to a museum with a good camera. I don't think measurements are really necessary... only thing that matters is proportions and you can eyeball that easy enough. I've typically found plenty of photo reference just with google search for most things.

    Anyway, I'll be bookmarking this thread and adding anything I find.
     
  10. yoonitee

    yoonitee

    Joined:
    Jun 27, 2013
    Posts:
    2,363
    Well i guess at least I've established a gap in the market for animal skeletal models.

    Pity I cant find a database in which to crunch all the numbers. Maybe i should start my own...
     
  11. BIGTIMEMASTER

    BIGTIMEMASTER

    Joined:
    Jun 1, 2017
    Posts:
    5,181
    If you don't mind my asking, what do you need "all the numbers" for?
     
  12. May I ask you why skeletons? Animal-zombie game? :) Because you don't need bone-measurement for modelling animals (at least not this precise).

    BTW:
    Animal Bone Metrical Archive Project (ABMAP)
    http://archaeologydataservice.ac.uk/archives/view/abmap/search.cfm
    (IDK if it works or if they have good data, it was google-fu)
     
    Last edited by a moderator: Oct 5, 2018
  13. yoonitee

    yoonitee

    Joined:
    Jun 27, 2013
    Posts:
    2,363
    I wanted to try an create an encyclopedia of animal anatomy in 3d. Also i wanted to populate a world with 100 types of animals without having to model each individual one. A pipe dream maybe. (Yes they could be zombie animals! Or maybe the player is a robot with x-ray vision!)

    I had this vision of being able to morph any skeleton into any other. Such as a lizard skeleton into an ostrich.

    Instead maybe i can do something like spore creature creator and have people make their own animals from templates.
     
    Antypodish, neoshaman, Ryiah and 2 others like this.
  14. BIGTIMEMASTER

    BIGTIMEMASTER

    Joined:
    Jun 1, 2017
    Posts:
    5,181
    Scanned 3d assets in their raw state wouldn't be suitable for a game engine. You'd still have to do a lot of work.

    It's a cool idea, but I don't think it hinges on having a library of animal skeleton scans available. Firstly, that would be enormously expensive -- or not something you could use commercially. Secondly, as I said, a scanned model is just a jumble of polygon soup that will have to be retopologized, unwrapped, all of that, to work in a game engine. Third, unless you just want animal skeletons running around... you still have to make all the models. Skeletons are great for ensuring you model them accurately though.

    Do you have zbrush? You can take that jaguar skeleton model I linked, use that as a base to build a big cat from, and then use something like a multi-pass fur shader (a couple on asset store) to get some decent looking fur without a ton of time investment. Benefit of starting from the skeleton is that it will be a lot easier to ensure your proportions are spot on, and you'll develop understanding of how the animal is built and thus will move and deform by piecing it together muscle by muscle.
     
    Last edited: Oct 7, 2018
  15. Kiwasi

    Kiwasi

    Joined:
    Dec 5, 2013
    Posts:
    16,860
    I like the idea. It could be fun. Pretty much all tetrapods have very similar bone structures. I'm sure someone clever could build a cool toy that merges between them.
     
  16. yoonitee

    yoonitee

    Joined:
    Jun 27, 2013
    Posts:
    2,363
    Yes. I mean Im not sure what the point of it would be exactly. But in my mind I am telling it "show me a cat" and it shows an animated cat. Then I say "show me a frog" and the cat morphs into a frog.

    I guess it would make an interesting shape-shifter game.
     
  17. Put a database next to it with scientific name, origin, stories, etc and you have an excellent educational VR-experience.
     
  18. hippocoder

    hippocoder

    Digital Ape

    Joined:
    Apr 11, 2010
    Posts:
    29,723
    I suggest @yoonitee go right ahead and do this project with made up bone lengths to start. You can just maker an editor and let people put their own bone sizes in. Good luck :)
     
    Kiwasi likes this.
  19. yoonitee

    yoonitee

    Joined:
    Jun 27, 2013
    Posts:
    2,363
    Yes, that's a good idea.
    Well I was thinking of using a "generic" cat model for lion, tiger panther, house cat etc. Except it turns out these animals are pretty different from each other. And a "generic" mammal... what would that be?

    But yes, a few basic model types like bird, lizard, horse, plus an editor would be kind of interesting on its own. Kind of like a wiki for skeletons.
     
  20. zombiegorilla

    zombiegorilla

    Moderator

    Joined:
    May 8, 2012
    Posts:
    9,042
    I'm sure such a thing exists, probably multiple ones. I know there is extensive morphological skeletal data on dinosaurs. (used for predicting and matching new skeletons and for things like locomotion). Through the data isn't probably readily available (or all encompassing in a single place) or in a directly useable format/structure. Since stuff like http://eskeletons.org exists, the data exists. If you are serious about it, just contact a few Comparative Vertebrate Anatomy departs (or related), and start asking questions. They'll be able to point you in the right direction.
     
  21. yoonitee

    yoonitee

    Joined:
    Jun 27, 2013
    Posts:
    2,363
    You would think. But I would imagine scientists do pretty similar things to what they used to do before computers. Looking up key measurements in tables. And using their expertise. I cant really think in what circumstances a zoologist would suddenly think to create a 3d skeletal model of an aardvark. They probably would have neither the knowhow or the finance. Computer modellers are expensive people!

    But you are right I should send out some emails.
     
  22. BIGTIMEMASTER

    BIGTIMEMASTER

    Joined:
    Jun 1, 2017
    Posts:
    5,181
    You probably won't find the diameter and length of an aardvark femur very easily, but I bet you can find the average length and height and weight of an aardvark. And I bet you can find plenty of aardvark skeleton picutres. So from that, you should have enough to build a realistic enough aardvark skeleton. Bone sizes vary a lot anyway, so as long as you have the proportions right -- which is just a matter of looking at the thing carefully, maybe using a grid in photoshop to break it into pieces -- you should be good.

    Why not start with a housecat to jaguar transition as proof of concept? You can use that free jaguar skeleton, edit it slightly to produce a housecat skeleton, and then you've just go to figure out the morph animation part.
     
  23. zombiegorilla

    zombiegorilla

    Moderator

    Joined:
    May 8, 2012
    Posts:
    9,042
    There are tons of uses, evolutionary, comparative, population diversity, developmental, and locomotion to name a few. The amount of data and information collected for this kind of work is orders of magnitude in scale beyond simple game use. The challenge would be in simplifying and just extracting the information needed for simple game physics. That and the fact that areas of research tend to be highly focused. There probably isn't a single source that covers "all animals", there are probably single sources for things like equines or felines or birds of south america or whatever. Evolutionary studies would probably be a good source, as they look at change over time, they will probably have higher sample rates. (or possibly lower, like with dinos).
     
  24. zombiegorilla

    zombiegorilla

    Moderator

    Joined:
    May 8, 2012
    Posts:
    9,042
    Maximum Femur Length (mm): 193.9 ± 17.1 (n = 43)
    Proximal Mediolateral Breadth (mm): 67.7 ± 7.4 (n = 43)
    Distal Mediolateral Breadth (mm): 58.3 ± 4.7 (n = 43)
    Distal Dorsoventral Breadth (mm): 58.1 ± 4.3 (n = 40)
     
  25. BIGTIMEMASTER

    BIGTIMEMASTER

    Joined:
    Jun 1, 2017
    Posts:
    5,181
    Haha. There you go. Now we expect a scientifically accurate zootopia!

    But, have fun modeling that. I think you'd need to use a CAD program or something to work with such precision. At least in my short modeling experience, the apps you typically use for game models aren't really designed to work with precise measurements like that. Maybe Fusion 360? I don't know much about that but it's popular for making guns from blueprint specs. But the common complaint with that program is that it outputs models that still have to be retopo'd in another program. It's not clean, low-poly game geo.
     
  26. zombiegorilla

    zombiegorilla

    Moderator

    Joined:
    May 8, 2012
    Posts:
    9,042
    You don't need to model bones. And if you did, you would just variant data. For game use, all you need it bone length, and joint information (rotational limits, etc).

    But that isn't the point. All the data is out there. Lack of data isn't the problem by a long-shot. Whether or not it is practical or useful on any level is the issue. 99.9999% of the time it is not. There pretty much isn't any challenge to doing it, the challenge is why would you do it? (how many animals does a game really need, how accurate do they really need to be?) There are tons of existing commercially available models for virtually everything. Creating "procedural animals" from existing data sources is pretty much a magic button solution (OMG! ProcGen!!!). It's impractical, redundant, and unnecessary. The technical term is procrastination. Like the old saying goes... if you have to ask...
     
  27. Kiwasi

    Kiwasi

    Joined:
    Dec 5, 2013
    Posts:
    16,860
    Museums. These guys are always looking for new ways to engage an audience. A digital experience where you could explore the various skeletons in the collections would be cool. Especially if you set the system up to follow the natural evolutionary pathways between animals.
     
    zombiegorilla and BIGTIMEMASTER like this.
  28. neoshaman

    neoshaman

    Joined:
    Feb 11, 2011
    Posts:
    6,492
    Well I'm eventually doing just that, at least as soon as I stope being bothered by the convex hull code, which exist to pose the basis of just that. I already put lot's of thought about it, now I just need to properly implement it.
     
    yoonitee likes this.