Another Big Hero 6 Study... Last One!

Okay, so I know I've done so much Big Hero 6 study. This is the last one, I saw a really cool weight shift and natural looking movement that I wanted to try and copy.

So I was going for a direct copy. I liked how he shifts his weight from one side to another. In the time-lapse you'll see that I scale up his hands, it was only to account for the gloves Hiro wears in the movie. Also this is the first time I've animated eyes, I wasn't sure what exactly to do. I figured eyes don't really slow-in/slow-out, so I made them dart and put it into linear.

The reference is from this lovely scene. I also had to account for the fact that Hiro in the movie is standing on something that isn't flat, but it's pretty close.

The head is a bit snappy and the arms need some overlap. Maybe I'll add those later. The things you notice after you upload!

What I Learned
The biggest takeaway from this is to break up movement. At one point I delete keys because I confused myself and added so many. One thing I learned is the you can find your poses and key those in, then adjust timing and offsetting afterwards.
I'm also teaching myself facial animation right now, just started with the eyes. So any tips on that would be awesome.

The time-lapse is three different recordings. I started one at my coffee table, so that's why the ratio is 16x10. The last recording is on my Cintiq screen so that one is at 16x9.

Big Hero 6 Animation Study Part 2!

Another study early this week, I won't do any more studies until my next assignment is completely finished. I feel like I've learned a lot from these and I'll keep doing them, but I want to apply what I've learned to my assignment now.

My study is from Big Hero 6 again, here's my final video:

I wanted to do a stillness study. What happens in animation when someone is more or less still. The arm rotation really stood out but what else goes on?

Not much apparently, mostly just offsetting keys... But here's the reference I used!

Mostly what happens on the still part of the shot is arm movement. I also made his torso straighten and uncompress from the previous poses, but that doesn't take long. Something else that I started implementing was something my mentor Jon said about rotating arms, don't just move around the rotation selection in all axis, it could cause some problems. The problems can be fixed, but they're a pain. So going forward I've been trying to use a specific axis on the rotation handle. Another big thing I learned is to look at your curves. At one point while animating there were some weird looking movements. I figured to check in the graph editor and there it was! Weird curve problems.

Here's the time-lapse video of the animation process. It took around 42 minutes to finish but the video is sped up to be around 15 minutes. I hope you enjoy and learn something!


I won't post for a little while while I focus on animating my own scene, so see you next week!

Also, here's the .mb file and the eleven rig download: https://drive.google.com/open?id=0B01q5eoTgo3ZQVUyYmxXNGtmMlE Extract the eleven rig and open the .mb file, when it tells you to relocate the rig go to: ElevenRig_v1.2/Scenes/ElevenRig_V1.2.ma

Workflow

Jon's Workflow
*I took these notes during class, so it's kind of a mess!

Visualization. When getting a dialogue shot, put it on repeat and listen to it over and over again for about an hour or so. Try to find different shot ideas in your head. If you come across something you like, take a pause and draw thumbnails. Keep them fast and efficient, get the essence of the pose and idea. You want to keep momentum going, so keep it brief. These thumbnails will allow you to pitch the idea to somebody. Now clear that all out of your brain and start again. Do this at least three times with completely different ideas, don't keep them too similar. Don't get married to an idea. Sometimes people could have feedback that could add to the idea. If you're too closed off on one of your ideas, you could be limited in how accepting you are to ideas. Believe in yourself through ideas, don't feel like you're doing what somebody else wants to please them or because you're not confident in your own idea. Only add them if you want to, you're the one doing the shot.

Reference. Using two cameras can be useful. Recording a few takes from multiple angles gives you a lot of variety to sift through. Don't skip on this phase, it's very important. Hit ideas and poses, but continue other variations. Try not to redo the same thing every take, try more. Get out of your comfort zone. If your reference looks robot-y, your animation will too. Get loose, study acting, and practice! You'll get better over time.

Blocking. Use stepped if you can, get the key ideas in. Now you can play around with how you move through poses. Showing really rough things is okay at certain studios. It works best to get core ideas into stepped. Don't over-breakdown the shot in stepped, it's easier to get the finished result in spline where everything is smooth. Instead of taking a lot of time figuring out in reference, put rough poses in the blocking to pitch with. You can easily see what's going on in the viewport than what you're doing in person. If you overcomplicate, though, it could lead to deleting frames in the future, which results in a waste of time when you added those frames. Keep it simple.

Spline. Ignore the arms. Focus on the hips/root in relation to the footfalls. Look at the pacing on ones. Think about repetition in shots, break up evenness, and involve variation. Watch root from start to finish, if it's 90% right, the rest of the animation will be easier. Now look into offsetting keys, overlapping actions, and follow-through.

Polish. Make sure the root works and look into all of the details. Arc track your points. Brows and eyes are done before the rest of the face because it's much easier to handle. Work through the facial animation in spline, stepped will look bad and will be harder overall to figure it all out.