Adán - VIZ Blog

Not really, but it looks like it.  Either that, or she’s got her head out the car window.

Here are a few shots of the beginning phase of modeling Sadie.  I was able to get her to pose for some pictures to use as reference images and that has helped out a lot.  Her ears are weird because I figured it’d be easier to skin and rig them while they’re easy-to-get-to instead of when they’re folded down.

Again, this is just part one of the modeling.  I have to convert to polys and stitch everything together to be one piece of geometry.  Then I might take it into Zbrush and play with things there, otherwise I’ll just refine details using either a smooth proxy or converting back and forth between polys and sub-d’s (there is an easy way that I found out from the tutorial previously mentioned…).

For my final, I want to revisit what I started working on at the beginning of the semester which was a quadruped rig (via my dog, Sadie).  This time, however, it’s going to work right.

I’ve been watching different tutorials on modeling and asked around and I think I’ve come up with a way to model that will work for me.  It’s kind of a weird workaround, but every person is different and this seems to work best for me.  The tutorial that helped me along was from a recently discovered site: simplymaya.com  (to find the tutorial, just google: simply maya cartoon dog).
For the final, I am going to create a quadruped rig, skin it, and create a realistic looking walk cycle.  Hopefully a run cycle too if I can get to that.  This will all be from my test subject: Sadie.  She’s a good sport.  Also, time pending, I would like to recreate some of her performance from some of the tricks she knows.  She’s a very expressive dog and it’s interesting to watch her.  It’d be cool if I can recreate that.

So now, off to modeling!

[placeholder]

Here is a test animation that I made that has the bones go through a “calisthenics”  motion to which I skinned the overlaying geometry.  The animation made it easier to check things along the way instead of having to move something, fix the skin, then undo to change it back.  Most of the motion looks right, but there are some problems that I faced along the way that were mainly due to my bad modeling skills and not necessarily the skinning process.  I think I spent about 85% of this project modeling…and I still didn’t get it right.

[placeholder]

This is the cycle that I created for the Deinonychus.  I decided not to go with a range of motion thing for the arms and instead try and create a cycle that more represents what the character would normally do.  Basically, I animated how the arms would look during a walk cycle.

There’s something a bit off about it and after my review it turns out that the arms don’t really show the wave of energy that happens when the dinosaur would take a step.  As in, when the foot falls, the hand isn’t going to fall down at the same time, it’s going to be a little delayed…

For this next assignment, I am supposed to rig the shoulders/arms/hands OR hips/legs/feet of a biped (orthograde) creature. In class, I asked if we could do something like a dinosaur as an alternative to something like a human and was given approval. I just figured that most people would just start rigging up humanoid creatures and I personally don’t want to do that (nothing against that though!). A lot of the dinosaurs that walk on two legs are actually pronograde (spine parallel to the ground). This makes things even more interesting…for later on down the road, if I want.

So I decided to not rig up a T-Rex, because when people think of dinosaurs, that immediately comes to mind. Instead, I chose the second most famous carnivore, in my mind at least, the Velociraptor. When I think of a “raptor,” I think of what I have seen in the Jurassic Park movies, and that’s not necessarily a true Velociraptor. It was a Velociraptor antirrhopus (at the time) which is actually a Deinonychus. Either way, this dinosaur walks on its toes (digitigrade) like a dog and I want to try and get a better approach to something that walks like that (in regard to my previous assignment).

I searched online for reference skeletal images of a Deinonychus and came up with very little. In my searches, I came across a professor from the University of Maryland, Dr. Thomas R. Holtz, a vertebrate paleontologist. I sent him an email asking if he knew of any references that would help me with this project and he quickly got back to me with the names of two other professors: Dr. Phil Senter at Fayetteville State University and Dr. Alan Gishlick at Gustavus Adolphus College. Dr. Holtz told me that both of them have done research on articulation of the forelimb of Deinonychus and other dinosaurs, trying to answer with physical models the sort of issues I want to work on with visualization. I have yet to hear back from them but I sent them emails late Friday (weekend…).

This project has started to become something more than what I was initially expecting, but I am really interested in this whole process!

For this assignment, I looked at footage from the movie The Grinch (the Jim Carrey version). There is a particular scene where the Grinch keeps trying to scare Cindy Lou Who, but to no avail. I tried to mimic the types of shots that were used, primarily the sequence: both characters, stationary character, action character, both characters, action character, high angle.

I kept Sadie’s action the same except for the end when I have her look up at the camera in the same manner as the Grinch when he has a little aside after his failed attempts.

I chose this movie clip because it mimics Sadie’s failed attempts to get the toy down. That, and I love that movie.

[animation]

…I think.  I took my initial idea one step further in Flash by trying to make it look less like MoCap.  I mainly used shape tweens to show the movement of her body based on the timing that I concluded from the footage.  I tried to go piece by piece and try and figure out what it was that my eye was seeing.  I tried to leave out parts of her body that I ignored when looking at the footage and I think the effect is pretty good.  I left her tail kind of choppy because she swings it back and forth so hard and fast that it’s usually just a blur (except at the extremes).  I left just enough frames of her tail so where you know the movement without actually having to see all of it.

Click To Play

I recaptured the footage using two camera angles and picked one to use as my source footage. From there, I wanted to dissect Sadie’s motion by taking the video into flash and creating something that looked like a mixture of rotoscoping and motion capture…

Click To Play

Click To Play

For this assignment I used an animal resource that I have on hand 24 hours a day–my dog, Sadie. I filmed her doing various tricks and narrowed it down to just one video of her jumping up to catch a rope hanging from my chin-up bar.

Here are some YouTube links to the initial reference footage:
http://www.youtube.com/watch?v=lvi-QXyVQcI
http://www.youtube.com/watch?v=CZObaD507PQ
http://www.youtube.com/watch?v=jiMR9xAAWSM
http://www.youtube.com/watch?v=ZxWOB82HId4
http://www.youtube.com/watch?v=r455WkXilS8

and if you’re just interested in some of the things she can do (longer video, 2mins):
http://www.youtube.com/watch?v=IBcExTb7lds

I worked with Ganesh Rao on this assignment and our submission can be found on his page.

Soon I will post the process photos…