Saturday, September 22, 2012

KickSTARTED... Past Tense

Holy Houdini, we did it. As of 8:21 yesterday evening, our Kickstarter was officially a success, earning in excess of its $5,000 goal. I'd like to extend a huge thanks to everybody who contributed or shared our links amongst their web friends (and possibly web frenemies). It's going to mean some huge progress in the upcoming months, shooting some stuff I'll be quite excited to blog about. To celebrate our good fortune, how about a behind the scenes video featuring a conversation with the ever-engaging Drew More. In certain circles referred to as the Orson Welles of visual effects.

And just in case any of you missed it, last week we also uploaded a detailed documentary about shooting our promo clips.

A fair warning, both videos are quite lengthy, and may only appeal to those with an intense interest in the indie-filmmaking process. Or... AN indie filmmaking process, anyway. Not sure anybody else does it like this...

Saturday, September 1, 2012

We've All Gotta Break Down

Yes, we ARE still begging for money on Kickstarter. And if you checked the update Sean posted today, then you've already seen how I spent my past couple of days. But since it's been kinda dead here and I've got nothing much else to write about, I'll go ahead and give you a breakdown. We'll forego the image-based journal entry for this one, in favor of a video instead. Give this bad boy a watch, if you like, and I'll explain what you're seeing a little bit.

Basically, Sean and I both wanted to have something new and at least somewhat interesting as a thank you to the donors we already had, and maybe an enticement to some new ones. I decided on this little moment because it's sort of a nice emotional beat, which is a side we haven't really shown of our project yet. I also needed to make sure that it was something I could accomplish in a couple days with the resources I have available to me. The end product obviously isn't final quality, but it conveys the spirit of the scene fairly effectively.

So anyway, what've we got here? First you'll see the original raw footage. As is apparent, that is not at all a forest they're in, despite the fact that there's plenty of green. So obviously, as with all our Dorothy footage, compositing is required to bring it anywhere near a serviceable state. You'll also notice that as with much of the Dorothy footage, it was shot handheld on a shoulder rig. So the first step was to run a 2D track on those markers and key out the green.

Next up is the entirety of the ADR session for this particular shot. The number I call out at the beginning is the file number on the raw shot we're matching to, so I know where this audio is meant to go when we're editing. You've seen snippets of this process in the previous entries, but I thought it was interesting to show how we played with the performance after the fact. We did multiple takes, and in addition to being different from each other, they're also quite different from what she did on set.

Then we have the most critical element of the shot, our CGI Nick Chopper. Knowing I only had a few days, this was always going to be the hard part. The existing Woodman rig is purpose-built for the last shot of the Concept scene, he hasn't been fully rigged yet. So I had to do a quick five-bone rig and some morph targets on the eyebrows for my purposes here. The animation was where I labored most. I didn't have the kind of time I'd like to have perfected it, but it nevertheless went through multiple tweaks and iterations.

The lip sync is of special note, I think. This is the first time we've seen Nick doing any. Well, not entirely. I'd done a bit of goofing off a while back to experiment with his facial capabilities. And it wasn't very successful. I made the mistake of animating his jaw flaps to match the human lip movement of the performance, and it didn't work right, given the fact that he has no articulation. This time I filmed some reference of my hand doing Muppet-flaps to the dialogue, and matched his jaw flaps to that. This worked much better than my previous test.
You're telling me something about this was ill-advised? I don't believe it...
The final render of Woodman was where the time crunch was most felt. I did some experimentation with a different light set-up, trying to cast some leaf shadows on him, and it looked quite cool. The downside was that on my computer it was going to take 27 days to render. I pared everything way down to the bare minimum and got it to render in 27 hours instead.

So with our live-action characters keyed, Nick animated, rendered, and tracked in, all that remained was the background. Again, time wouldn't allow for an actual CG background to be assembled or rendered, so I popped out into the woods and found a suitable location that would work for my staging and lighting. I had to Photoshop out the evidence of snowmobilers, and Photoshop in the evidence of yellow-brick-layers, and after that it was very easy to drop it into the composite.

The final thing you see there was my first pass of the shot. I was originally considering a split diopter sort of effect. However once I saw it rendered I realized I wasn't fully in love with it, and in the context of a single-shot clip it would probably end up being little more than a distracting element. I quickly ditched it in favor of just a deeper focus.

So there you have it, in lengthy, gratuitous detail. Everything it takes to make a semi-presentable version of a single shot in three days. An actual 3D track and CG environment would do a lot to liven up the background and Woodman's eyes didn't come out quite right for various reasons, but I think it does alright for what it is. And at least for Sean and I it served as a somewhat relieving proof-of-concept for our thoughts about Woodman's performance.