Salty's Animation Bootcamp - Principles of Construction and Deconstruction Pt.2 (Patreon)

Welcome back!

We started off by giving you an example of how using simplified shapes, squares and rectangles in our example, gives a framework for accuracy and consistency when animating a head.

I'd like to expand on these principles and show you how you can use it for broader and more complex scenes.

Suppose you were tasked with animating a person doing a complicated movement like martial arts or dancing. How would you start the process? Personally, I like to reference actual video of people doing these actions. 

I watched a few video and sketched out a stick figure animation from the side. Really I was just focusing here on the spacial relationship between the arms legs and feet, not worrying too much about details. 

Try to only focus on the important keyframes like when the person fully extends the kick or when he lands. 

Here's the reference animation:

 Depending on your skill level, you may not be able to do this very accurately or efficiently. I personally don't see a problem with tracing, especially if you are starting out.  I didn't trace this - not because I didn't want to upset the art gods, but because it was way faster to do it by eye. You do you.

Now that we have an idea of the movement we want from a side, we want to place this action in a world and make is look believable.

I chose a relatively complicated background. It has a distortion on the lens that I made in a 3d program simply to illustrate how effective our breakdowns will be.

If you are new to animation, this can look pretty daunting, but it really isn't. You already did the hard part, which was capturing the motion in a stick figure.

Ok so where do we begin? First, I say we think about where the character will move with respect to the camera. I decided on making the person move closer to the camera during the kicks. So what we need now is a bounding box: a box where most of our actions will be.

I made this 3d background specifically with a grid to help with laying out your animation correctly.

From here I decided to show you what I thought the animation should look like without using and tools to guide me through the process. 

I got the idea down, I guess. I don't really think this up to my standards so lets try to clean this up using guidelines. Lets make a smaller bounding box where each individual frame will occur. I made the first and the last frame, then just filled in by inbetweening.

Now that we have the bounding box for every single frame, we can now reference our initial animation to make a more accurate one with the box we drew.

Notice I deliberately made the reference animation done from the side. I had another go at the first pass, but now with a bounding box and a reference to help me.

I would say comparatively this pass is much better than the one I did just using just my eye. With the bounding box, I feel like it's much clearer where the character should be at any given moment.

 Here's the pass without clutter.

Now that we have our first pass, let's try to put a character in.

I made a random one.

Now let's try putting meat on the bones.

The second pass isn't looking too bad. I personally didn't like a few things that were going on with the proportions and the position of some of the frames so I did a final pass to clean it up.

The same scene with the background.

The last pass of the animation is much more believable, I would be really confident in taking this to the line work/inking stage. I think the only thing I would do from here is to just adjust some of the timing and spacing, but that's for another topic.

Let me know if I left something out or if it needs to be clarified.