The Advanced Art of Stop-Motion Animation: Visual Effects - Part 1
Many shots are done in a simple split-screen technique, where live action and stop-motion are shot as separate scenes and brought together into one shot. This can be done very easily in any non-linear editing program by applying a mask with an alpha channel into one of the scenes, and then layering them together in Premiere or After Effects. In this situation, the split-screen matte line still acts as a division where the two elements should not cross over each other (Figures 9.14 and 9.15).
Other shots require a little more work and planning in the compositing and layering to bring Barry into interaction with the live-action world. Here, Rich himself describes the steps he takes to accomplish this:
I start by locking the camera down, and shoot the live-action video with markers so the actors know where Barry is going to be when looking at him or following him as he moves. I also make a rough note about how long things take and what new improv comes out of the shoot so that I know where I need Barry to move, react, and look. After the live-action video is done, I use a remote to capture frames of Barry moving around with a clean background behind him. I also take one or two frames of the clean backplate with no Barry or actors, in case I need it for any holes and to mask bad reflections or unwanted shadows.
For a shot in Episode 1 where Barry comes out from under the bed and rolls in front of Frank, three layers are needed to make this comp work:
1. Stop-motion layer with Barry animated and saved out as a high-res MOV file the same frame rate as my live-action plate. In this case, it was NTSC 29.97 (Figure 9.16).
2. Live-action video layer with Frank, shot using NTSC 29.97 frames per second, in standard definition (Figure 9.17).
3. Clean background plate in case it’s needed (Figure 9.18).
I import and/or capture the stop-motion and video layers into my editing program in this same order, with stop-motion on top. I use a temporary “garbage matte” (drawing a rough matte around the general area where Barry is) on my stop-motion layer so that I can see the video layer underneath. If you can’t make a temp matte, another method is to reduce the transparency. The key is to be able to see both layers so that you can match them up for your final edit before compositing them together. This is the most important step, and you need to lock down the edit in this stage because the last thing you want to do is go back and make changes. It's too much work to do that. Each layer is edited and timed out, the temp matte is removed, and stop-motion and video layers are exported as uncompressed files to my compositing software.
Then, I import the uncompressed files into compositing software the same way, with layers arranged top to bottom. I add a 2-pop* 1 second before and after each clip to help ensure that they are lined up.
[*Author’s note: A 2-pop is a sound tone one frame in duration that is typically placed 2 seconds before the exact start of a program for cueing purposes.]