Open-Source Compositing in Blender
Linux is bursting with multimedia potential—at least, that's the impression one gets from the plethora of multimedia-oriented distributions that have surfaced in recent years. DeMuDi, Planet CCRMA, Ubuntu Studio, 64 Studio and the list continues on ad infinitum. However, for many years now, the term multimedia proved deceptive. GIMP and Inkscape and other graphics tools meant that 2-D graphics were covered, and the astounding variety of audio tools available with real-time priority meant that users' needs for recording and audio processing were met. Video tools lagged behind, because video processing is both more difficult and more prone to patent encumbrance. In the last few years, things have begun to catch up to the point where it's feasible to create films or cartoons from concept through execution using only Linux tools.
Elephants Dream, one such cartoon, was the foundation for a major breakthrough in open-source video. Financed by presales of a then-unseen cartoon, Elephants Dream was a strategy for raising money to advance the development of and raise awareness for the open-source 3-D suite Blender (www.blender.org). In order to accomplish this goal, the creators had to develop something that never had been available before: an open-source compositor.
Compositing is the art of taking multiple image sources—whether from 3-D, vector graphics, photographs, video or procedurals—and marrying them together to create a seamless, integrated image. A good compositing program provides the means to access all the mathematical functions available in the image processing universe, and a good artist needs to be able to get down into the guts of an image from time to time, below the interface, and tweak it directly with mathematical functions.
Because of Linux's continuing adoption in post houses, several high-end compositing systems, such as Shake, D2 Nuke and Eyeon Fusion, have been available for years now, but the prices run up to thousands of dollars per seat with extra costs for maintenance and render-node licenses. For those with smaller budgets, Linux compositing has been perpetually out of reach, unless one has the patience to do hand scripting in ImageMagick, which is far more technical than most artists care to get and generally requires the addition of a hacker to the pipeline (another cost point), or work frame by frame in GIMP, which is laborious and not worth the effort for any but the smallest projects.
Consequently, the only budget-friendly solutions for a small studio has been Adobe's After Effects or Apple's Motion, which means adding a Windows or Mac OS machine to the pipeline. Both are very capable, useful tools that produce professional results, but neither are open source.
The two classes of compositors are built around two divergent interface paradigms, which dramatically effect work flow. The first of these paradigms is the Photoshop paradigm. It shows up most prominently in After Effects, and in a modified form in Apple Motion, and works by marrying the interface conventions of Photoshop with a basic multitrack editor interface. In this paradigm, composites are achieved using layers of images built atop each other, with each layer being operated upon by its own effects stack. The main advantages of this paradigm are the speed of work for simple and moderately complex effects and the ease of navigation for new users who already are familiar with Photoshop (Figure 1).
The second paradigm, the “node-based” paradigm, is the one that appears in the high-end professional compositors. It works by chaining together various image functions to create complex effects. Image functions are mathematical transforms applied to an image to change it in one way or another, and they reside at the base of anything one does in GIMP or ImageMagick or in video compositing. These functions are encapsulated in the interface by nodes. A node works a bit like a guitar pedal—it accepts inputs and generates outputs, and those outputs can be routed to an infinite number of other nodes. Thus, in a node-based compositor, one uses the node chains to accomplish one's goal, and there typically are two types of nodes from which to choose. One type is the familiar, user-friendly prepackaged effects plugins, such as one would find in the Photoshop universe. The other type is a set of mathematical interfaces that allow you to build custom effects yourself. This has the disadvantage of being far more visually complex and, for some people, much harder to learn. However, for that steeper learning curve, the artist gets a much more versatile work flow, which is better suited to doing highly complex work. Node-based compositors available for Linux include: Shake (now defunct), Eyeon Fusion, D2 Nuke (formerly of Digital Domain, now owned by the Foundry) and Blender (Figure 2).
Blender itself has long had a rudimentary track-based compositing system, which has received a lot of attention since Elephants Dream and has become quite useful both as a video editor and a compositor. Alas, because its primary purpose is video editing, it lacks the ability to nest compositions or layer effects as complexly as can After Effects and Motion, leaving it firmly in the quick-and-dirty category for compositing work.
However, with version 2.43, Blender introduced its node-based compositor, the jewel in the crown of the Elephants Dream improvements. Professional-level open-source compositing has arrived, and it's integrated with an otherwise very powerful 3-D content creation suite.
To demonstrate what it can do, let's walk through a fairly simple five-layer composite.
To get to the compositor, fire up Blender and change the main 3-D window to the nodes editor (Figure 3). Toggle on the Use Nodes button. Because Blender uses nodes for material and texture creation as well as compositing, you need to depress the picture icon. By default, a render scene and a composite node will pop up. In the case of this project, one of the elements I'm using is a 3-D scene in Blender, so I retain this node and assign it to my primary camera (Figure 4).
Next, I split the bottom view into two windows, and in the right-most pane, pull up the image editor window, where there is a list box that allows you to choose the output nodes from the compositor window. This is how you check your progress (Figure 5).
Next, I add a few more sources. In each case, pressing space brings up a menu that allows me to add nodes. I want three additional input nodes, and from each I assign the appropriate type. For this project, I'm working with two still images (the lens flare and the background photo) and one image sequence (the greenscreen clip, rendered out from DV to an image sequence to make it easier to work with).