Editing: Layering Video

Bill Davis
September 2003

Over this past spring school break, we took a family trip to the Grand Canyon. It's an amazing experience, in part because the canyon itself is a kind of map of the Earth's geological history. Dramatic horizontal bands etch the canyon walls with different colors. I remember laughing to myself and thinking, "I go on vacation and I end up staring at a colossal timeline."

Of course, my edit suite doesn't come with pine trees, fresh air, or planet-class scenery, but one of the coolest things about modern editing is that it allows video editors to make layered video.

Once upon a time (in the 70s and 80s), if you wanted to put more than two pictures on screen at one time, you needed to go into a big postproduction house and rent time on a special box, typically an ADO unit. ADO (Ampex Digital Optical) beasts were both costly and rare. So costly and rare in fact, that the standard way to incorporate multi-layered effects in a program was to finish everything else, then bring the ADO unit on-line and start racking up production bills so steep they were usually itemized in quarter-hour increments. In addition, every time you wanted to add a layer of video, you needed to add an additional channel of ADO and the rate you paid would multiply.

Today, your software editor, no matter how basic it seems, can run rings around yesterday's ADO units. Even inexpensive editing software can stack multiple dozens of layers of video right out of the box. Multi-layered video is rapidly becoming a standard and, in many cases, our audiences expect to see multi-layer video. Today, adding a layer of video to a project is as simple as adding another video track to your timeline and re-sizing, positioning and compositing the resulting visual images so that they occupy the appropriate screen real estate.

The problem comes when you start to understand that three images running on screen at the same time represent three times the shooting, three times the logging and three or more times the work. Still, the results can be undeniably powerful and appealing, so lets' walk through a typical multi-layer timeline.

As our example, let's use a Grand Canyon Trip with three layers of video and a title layer. Layer 1 is a scenic shot of the canyon. Layer 2 is a scene of white water rafting trip. Layer 3 is a shot of a family camping. And finally, we have our title graphic, in layer 4.

Nearly all software editors vertically stack tracks for compositing layers. Some stack overlay tracks above and others stack tracks below the main timeline, but the principles are the same. As you can see from the final composite, the canyon shot is the background with the other three layers stacked on top of it. In this particular editor, the background canyon layer sits in the lowest track on the screen, track 1. The rafting shot is on track 2. We stack the camping shot on track 3, and the title on Track 4. In order to reveal the two back layers, we'll temporarily disable tracks 3 and 4. Now we can see the full screen image of track 2, which completely obscures the canyon background. In order to make a nice looking composite of the camp, we'll crop and resize the image and place it on the lower right side of our canyon shot. Now we can see the canyon.

The rafting scene also needs to be smaller, so we re-enable layer 3 and open up the motion tab to resize it to 25%. Finally, we'll re-enable our title layer. Since it was created with transparency already in place around the text, it composites over the other three layers and, with some simple size and position adjustments, we have a four-layered title composite ready to introduce our audience to our vacation video.

• Digg This!
• del.icio.us
• Technorati
• StumbleUpon
• Reddit