You may think that pulling a matte refers to keying out the blue or green from an effects film shoot (Figure 3.1). True enough, but there are other types of mattes too. Even more common than blue-screen keys are high-contrast, or hi-con, mattes. You create these by maximizing the contrast of a particular channel or area of the image. There are other types of mattes possible as well, such as the elusive difference matte. Chapter 6, “Color Keying,” discusses pulling mattes in depth.

Figure 3.1. This split-screen image shows a blue-screen shoot (left) and the resulting matte. (All baseball images courtesy of Tim Fink Productions.)

Using an existing alpha channel sounds like a no-brainer, right? The source footage was created with an alpha channel (typically in a 3D animation program); just bring it in as is (Figure 3.2). No worries, done deal, right?

Figure 3.2. This close-up of a computer-generated baseball is split into color (left) and alpha channels.

That's theoretically true, except that After Effects mostly conceals an issue that is quite explicit in other similar software packages (such as Apple's Shake): the interpretation of the alpha channel. Is the edge premultiplied or not, and what do you do in either case? The questions are raised not only when you import the initial image (as you do in After Effects) but also when applying effects to layers with an alpha.

How does After Effects get around this, and is it a good or a bad thing for compositors? For the answer, see the “Alpha Channels and Premultiplication” section later in this chapter.

A mask, by contrast, is a vector shape that determines the opaque and transparent areas of an image (Figure 3.3). Masks are generally created by hand, although After Effects does include a provision to generate them automatically by examining the raster data of an image. You can use the Layer > Create Outlines command, but typically it's not too useful for anything besides stylized motion graphics effects.

Figure 3.3. In this split-screen view, you can see the garbage matte mask that was added to remove areas of the stage that didn't have a blue screen.

This chapter focuses instead on the basics of creating and combining masks. You can find a follow-up discussion in Chapter 7, “Rotoscoping and Paint,” which focuses specifically on rotoscoping, the art of animating selections over time.

Use Blending Modes Instead

It's even possible to composite without selections at all, instead using blending modes (Add, Multiply, Screen) to combine color channels mathematically, pixel by pixel, in ways that mimic how light and shadow play out in the world (Figure 3.4). You can also use selections combined with blending modes to get the best of both worlds.

Figure 3.4. Blending modes are the preferred way for compositing elements that are composed of light rather than matter, such as fire.

This chapter goes over the modes that are relevant to effects compositing and gets into the nitty-gritty of what these blending modes are actually doing to the pixel data as they combine it.

Many effects can be used to create transparency effects or to refine the ones you have. Section II discusses these in detail.

As mentioned with blending modes, there is no reason to think that you're barred from combining these techniques, using a garbage matte to clean up a color key or a hi-con matte to enhance the effect of a blending mode. This is very common in advanced effects work, where no two shots are exactly the same and where a single frame or clip can require a variety of approaches.

The real skills here are knowing which approach to apply for a given situation, knowing how to apply it, and knowing when to try something else, either to enhance or to replace what you already have. None of these techniques or approaches is particularly sophisticated by itself, but applying them properly requires a clear vision of what the shot needs and how to supply it.

A bitmap selection channel is one in which each pixel is either fully opaque or fully transparent. This is the type of selection you get if you use, say, the Magic Wand tool in Photoshop. You can feature or blur the resulting edge, but the default selection has no semi-transparent pixels.

This type of selection has its place—say, isolating pixels of one particular color range to change them—but it is not how nature works. A bitmap cannot describe a curve or angle smoothly, and even a straight line with no edge transparency looks unnatural in an image whose goal is realism (Figure 3.5).

Figure 3.5. Using 400% magnification, you can see clearly how ugly a curved or angled shape is if it can be described by only bitmap pixels, pixels that are either fully off or fully on.

But although it's easy enough to see that a bitmap edge does not occur in nature, it's hard to imagine that a feathered alpha does occur—hard objects don't have semi-transparent edges, do they? Of course not. Look at the edge of this page, or anything in the foreground of your field of vision. Do you see semi-transparent edge areas? No.

Not exactly, anyhow. It so happens that semi-transparent edge pixels are the best approximation we have digitally for nature (Figure 3.6), because they solve two problems in translating the world of objects to the world of pixels:

Figure 3.6. Ahh, that's better. With a proper alpha channel and a one-pixel feather adjustment, even at 400%, the edge does a much better job of describing a smooth, soft curve.

Say what? The first point is easy enough to see and appreciate, but the second one is a doozy, an observation that can be traced back a century to Einstein's annus mirabilus, in which he demonstrated how mass and light are interrelated. Light is bent—slightly—by any amount of mass that it passes. The greater the mass, the higher the amount of bending, right up to the extreme of black holes from which light cannot escape. With most objects, your eye sees a slight, subtle effect.

Geek Alert: The Compositing Formula

When you layer a raster image with semi-transparent alpha over an opaque background image, here's what happens: The foreground pixel values are multiplied by the percentage of transparency, which, if not fully opaque, reduces their value. The background pixels are multiplied by the percentage of opacity (the inverse of opacity), and the two values are added together to produce the composite. Expressed as a formula, it looks like

• (Fg * A) + ((1-A)*Bg) = Comp

With real RGB pixel data of R: 185, G: 144, B: 207 in the foreground and R: 80, G: 94, B: 47 in the background, calculating one edge pixel only might look like

• [(185, 144, 207) x .6] + [.4 x (80, 94, 47)] = (143, 124, 143)

The result is a weighted blend between the brightness of the foreground and the darker background.

Other effects compositing programs, such as Shake, do not take this operation quite as much for granted the way that After Effects and Photoshop do. You can't simply drag one image over another in a layer stack, you must apply an Over function to create this interaction. Is there a difference? Not until you add to the discussion the operations that go along with an Over, in particular premultiplication, which is detailed later in this chapter.

You can study this effect close-up in a digital photo with no compositing whatsoever (Figures 3.7a, b, and c). In the digital image, you can see areas at the edge of objects that become a wash of color combining the foreground and background. This is caused by the influence of the mass of that object itself on the light (color) coming from behind it.

Figure 3.7a, b, and c. Consider a simple digital photo that is not composited (3.7a). When you look closely, you can see that there is a natural softness around the contour of the head despite that it is in sharp focus (3.7b). The effect is even more evident at 400%, if you compare the sharpness (albeit pixilated) of the detail in the fur to the softness of the edge (3.7c). (Images courtesy of Matt Silverman.)

But we're not done with the surprises connected to basic image combination; the way that Opacity edits work in After Effects often takes people by surprise as well, although they can work with it for years without coming face to face with what is counterintuitive about it.

Say you have two identical layers, no alpha/transparency information for either layer. Set each layer to 50% Opacity, and the result should look exactly like either one of the layers at 100%, right?

Wrong (Figure 3.8)! If you've ever heard of Zeno's Paradox, that's something like how opacity is calculated in After Effects, but for good reason.

Figure 3.8. Two white solid layers sit one on the other, each at 50%, yet they do not add up to 100% (full) opacity. They add up to 75%, hence the 25% visibility of the checkerboard background.

Zeno was evidently violating the bounds of common sense with his observation. But a lead developer on the After Effects team once described the program's opacity calculations similarly: Imagine you have a light which is 1. You now have half the light showing through (0.5 * 1 = 0.5). Put a 50% transparent object in front of that. Half of the light shows through (0.5 * 0.5 = 0.25), and so on.

Hence, and invoking Zeno's Paradox, After Effects is mimicking how transparency behaves in the real world, which you can try yourself using sheets of vellum or other semi-transparent paper. (This is extra credit, of course!)

Premultiplication exists for one reason only: so that source images look nice, with realistic, anti-aliased edges, before they've been composited.

That's right, we have premultiplication just so that a matted object looks right against a black background when it comes out of, say, your 3D animation program. All premultiplication does is composite the foreground against the background, so that the edges and transparency blend as well into that solid color (typically black) as they would against the final background.

It's not the alpha channel that distinguishes a premultiplied from a straight alpha image, it is the color channels. When you ask After Effects to “guess” how to interpret the footage (on import, by choosing Guess in the Interpret Footage dialog, or using Ctrl+F/Cmd+F), it looks for a solid color background with edge pixels that seem to have had that color multiplied into them.

What does it mean to have the background multiplied into the edge pixels? Revisit “Geek Alert: The Compositing Formula,” but imagine the background value to be 0,0,0. There's your answer. The effect is to darken semi-transparent edge pixels if the background is black, to lighten them if it's white, and to really wreak havoc with them if it's any other color.

The close-ups in Figures 3.9a and b show a section of the same foreground image with the alpha interpreted properly and with it misinterpreted. A misinterpreted alpha either fails to remove the background color from the edge pixels, or removes color that should actually be present.

Figure 3.9a and b. Motion Blur clearly reveals the sins of improper premultiply settings. Although there are dark areas in the blur of the properly interpreted foreground, they are consistent with the dark areas of the plane itself (3.9a). The improperly interpreted version has dark matting all around it, including in the areas of the canopy that should be translucent, and around the blur of the propeller (3.9b). You suspect it's wrong simply because it looks bad around the edges.

Why should you care about premultiplication? First of all, because incorrectly managing your alpha channels can cause undesirable fringe artifacts. But just as important, you may find yourself in a situation where those artifacts seem to be presenting themselves although you've carefully managed the process; suddenly your composited elements have black edges around them. Your job depends on getting to the bottom of this. There are two basic ways it can happen:

Note

Most computer-generated images are premultiplied, unless the artist takes specific steps to counteract the process. The Video Output section of the Output Module Settings for items in the Render Queue includes a pull-down to specify whether you render with Straight or Premultiplied alpha; by default, it is set to Premultiplied (Figure 3.10).

Figure 3.10. The cursor points at the Color selection, determining whether After Effects renders footage with straight or premultiplied alpha.

Unfortunately, users who aren't confident enough to trace the underlying problem often end up resorting to all sorts of strange machinations to try to get rid of the black edge; re-rendering the element against a different color, choking the alpha matte even though it comes from a 3D program and is accurate, and possibly even more frightening and desperate maneuvers.

Most people's ace in the hole if they don't really understand pre-multiplication is the Guess feature that sits next to the premultiplication settings in the Footage Settings dialog. There is also a preference that determines what happens when footage is imported with an alpha channel; if this is set to anything other than Ask User (the default) you may be importing files with alpha channels without knowing what is happening (Figure 3.11).

Figure 3.11. By default, the import preferences are set to ask the user how to interpret an alpha channel, which is a good thing, generally speaking, as a check against errors. If you're not sure about the appropriate setting, you can click on the Guess button, which typically can determine the type of channel. If it cannot guess confidently, it beeps.

Is Guess ever wrong? Yes, it can be, if the factors it expects in a premultiplied alpha are there in a straight image, or vice versa. Thus it is best not to automate this process, if only so that you're able to hear the beep. In the case of a premultiplied image, After Effects attempts to guess not only the setting but also the color of the background; generally this will be black or white, but watch out for situations where an artist has become creative and rendered against canary yellow or powder blue. For that reason, there is an eyedropper adjacent to the Matted with Color setting (Figure 3.11). When in doubt, look at your footage without the alpha applied: If you see a solid background that is neither pure black or white and After Effects isn't detecting it, use the eyedropper.

Note

To see footage that was imported with a premultiplied alpha displayed in straight alpha mode (with edge transparency removed) use this shortcut in the Layer/Composition window: Alt+Shift+4 (Option+Shift+4).

Fundamentally, though, as an effects compositor you need to be able to examine your images and spot the symptoms of a misinterpreted alpha: dark or bright fringing in the semi-opaque edges of your foreground.

The really gnarly fact is that premultiplication errors can be introduced within the After Effects pipeline. Usually this happens when you apply a selection as a track matte to footage (described in more detail toward the end of this chapter), and After Effects, of course, doesn't anticipate premultiplication. If the sequence was rendered with premultiplication but the alpha channel was created on a separate pass, as a separate file, for example, you would apply that alpha using a track matte. After Effects would have no idea that the image was premultiplied, however, because the color channels came in through the back door, as it were, without an alpha.

If you see fringing in your edges and need to solve it internally in After Effects, there is a tool to help: the Remove Color Matting effect (Figures 3.12a and b). This effect has one setting only, for background color, because all it does is apply the unpremultiply calculation (the antidote to premultiplication) in the same manner that it would be applied in the Footage Settings.

Figure 3.12a and b. In 3.12a, the plane against the white background had an alpha applied via a track matte without premultiplication removed, revealing white fringing, which is clear here against the black. With the color set to pure white, the Remove Color Matting effect corrects the problem (3.12b), but only once the color and track matte layers have been pre-composed.

Note

Remove Color Matting will not work properly on a layer with a track matte; because this is the main situation in which it's useful, be sure to pre-compose the layer and its track matte prior to applying Channel > Remove Color Matting.

There are three tools for creating masks: two shape tools and the Pen. To activate the Rectangular Mask tool, use the keyboard shortcut Q; press Q again to toggle the tool to the Elliptical Mask tool. If your mask doesn't conform well to a rectangle or an ellipse, you can draw it point by point with the Pen tool (G).

Whether you draw the mask in the Composition window or the Layer window is up to you. It is somewhere between difficult and impossible to draw or see a mask accurately in the Composition window if, say, the layer was offset and rotated in 3D space, but if you want to see the layer over its background, you need to have the Composition window open. With enough monitor space, an ideal compromise is to keep both windows open side by side, working in the Layer window and watching the Composition window for live updates.

Remember that your target shape doesn't have to be an ellipse to benefit from starting by drawing an ellipse. Indeed, if the shape calls for a perfectly circular curve on one side, you might do well to draw an ellipse (holding down Shift to make it perfectly circular) and then editing other sides of it with the Pen tool.

Note

If you need other common primitive shapes—a rectangle with rounded corners, a hexagon, or the like—you're best off drawing the shape in Adobe Illustrator, then copying and pasting it. This, however, will not work unless the preferences in Illustrator are set properly. Under the File Handling & Clipboard preference, choose AICB with Preserve Paths checked. This is not the default. You can also select and copy paths in Photoshop for use as masks.

When drawing a rectangle or ellipse

Highlighting a layer with a mask and pressing the M key twice in rapid succession reveals the full Mask options for that layer. Some tips for those are

Keyboard shortcuts are a big part of eliminating a lot of the fuss and bother that comes with masking in After Effects.

By default, the Pen tool draws Bezier curves, and in After Effects you can do everything you need to create a mask by keeping the Pen tool active throughout the mask edit.

I sometimes draw a Bezier mask first as straight lines only, clicking to place points at key transitions and corners. Once I've completed the basic mask shape, with the Pen tool still active, I can go back point by point and edit the shape, because I have instant access to all of the mask shortcuts:

Note

As with most selectable UI elements in After Effects, pressing F2 or Ctrl+Shift+A (Cmd+Shift+A) deselects the active mask. This is handy when you're done drawing one mask and want to create the next one without switching tools.

Only when you want to double-click to free transform the entire mask is it necessary to switch to the Selection tool, which you can do either by pressing V or by holding down Ctrl/Cmd while you double-click a vertex. The Ctrl/Cmd key toggles back and forth between the Pen and Selection tools, regardless of which one is currently active.

When combining masks that have semi-transparent areas, either because the opacity of the masks is less than 100% or, as in the examples shown here, because the edges are heavily feathered, you may not always want the densities to have a completely linear relationship in the overlapping areas. That's when Lighten and Darken modes come into play.

Figures 3.20a and b show the result of using each of these modes; they prevent mask densities from building up the way that they do with the other modes. No pixel within the combined masks will have a value that is not already represented in one of the overlapping masks; either the lighter or the darker of the two will be represented.

Figure 3.20a and b. With a lighten (3.20a) or a darken (3.20b) mask, the transparency values are chosen either from the mask set to this mode or those overlapping it, depending on which has the lighter (higher) or darker (lower) values.

Keep these in mind when combining multiple feathered or semi-transparent masks. Remember that mask modes apply to the layers above them, so these modes should be applied as each overlapping mask is added, not to the top mask in the stack.

A few features in After Effects exist specifically to help you manage multiple masks in a single layer.

One is Cycle Mask Colors, which is tucked away in the User Interface Colors section of Preferences. When Cycle Mask Colors is off, your masks are created in the same color (the mask itself and its swatch in the timeline) each time. To change it, you click the swatch. With Cycle Mask Colors checked on, which I recommend, the mask colors vary on their own.

Note

Don't forget that you can name masks by highlighting them in the timeline, pressing Return, and typing in a new name. The names don't come along when you copy and paste the layers, unfortunately, but they're still useful.

Also, when editing overlapping masks, you may find it helpful to use the context menu to lock and hide other masks. Context-click on the mask, and choose Lock Other Masks; now you can edit the active mask only. Context-click again, and choose Hide Locked Masks; this time your view is uncluttered as well.

You can set a temporal ease on a mask keyframe, but you have no corresponding spatial curves with masks. Each point travels in a straight line to its position in the following keyframe.

This means that to precisely mask an object traveling in an arc, you need to set a lot more keyframes than for an object traveling in a single direction.

The real bummer about mask keyframes is that you can't select a group of them and translate the mask; as soon as you move, rotate, or scale it, your selection snaps to the current keyframe only.

There is a workaround. You can duplicate the layer being masked and use it as an alpha track matte for an unmasked source of the same layer, in which case you're free to transform (or even motion track) the duplicate using the normal layer transforms. It's not a perfect solution by any means, but it is often useful and hardly anyone thinks of it.

You can freely copy mask shapes from one source (a different mask, a different keyframe in the same mask animation, Illustrator, or Photoshop) and paste them into an existing mask channel. You just have to follow some rather strict rules that depend on the situation:

In all cases, if the target layer is a different size or dimension from the source, the mask stretches to maintain its relationship to the layer boundaries.

In many situations, pasting Mask Shape keyframes blindly in this manner is not what you're after, but a rather obscure feature in the Layer window will help you. The Target pull-down along the bottom of the window has a unique function: When you choose an existing mask as the target, you can start drawing a new mask anywhere in frame and it will replace the shape in the target mask layer (Figure 3.21).

Figure 3.21. This pull-down menu at the bottom of the Layer window makes it easy for you to create a mask shape from scratch, replacing the shape in the target mask. If the target mask has keyframes, After Effects creates a new keyframe wherever you draw in a new shape for that mask.

First Vertex

Note

The Smart Mask Interpolation tool (available via a palette found in the Window menu) is designed to smooth transitions between two radically different shapes. It's useful for stylized motion graphics of detailed masks transitioning from one to the next, but less useful for effects masking and rotoscoping because it's a bit too automated. The result, while sometimes pleasing, doesn't usually obviate the need to more carefully keyframe an accurate mask transition.

When pasting in shapes or radically changing the existing mask by adding and deleting points, you may run into the issue of how the points line up. Hidden away in the mask context menu, and available only with a single vertex of the mask selected, is the Set First Vertex command. If you are having problems with mask points twisting around to a mismatched point during an interpolation, setting the First Vertex to two points that definitely correspond should help straighten things out. It also can be necessary for effects that rely on mask shapes, such as the Reshape tool (described in Chapter 7).

Add and Screen modes both brighten the image. Screen typically yields a subtler effect than Add, which results in brighter values overall.

Add mode is every bit as simple as it sounds; the formula is

where A is a pixel from the foreground layer and B is a background pixel (although they are obviously interchangeable in this formula). The result is clipped at 1 for normal 8- and 16-bit pixels; any pixels that add up to a value of more than 1 take the value of 1, full white.

Note

What is the difference between Add and Linear Dodge? The name! Open Photoshop and you'll notice it contains no blending mode called Add. Apparently Photoshop lacks the legal rights to use this term, hence Linear Dodge, which exists in After Effects only to match Photoshop.

This mode is used a lot; it brightens the overall image, but black in the foreground becomes transparent, effectively adding a value of 0 to the background. It is useful for laying fire and explosion elements shot in negative space (against black) into a scene, adding noise or grain to an element, or any other element that is made up of light and texture (Figures 3.31a, b, and c).

Figure 3.31a, b, and c. Add mode takes the source foreground element, the fire shot against a black background shown in 3.31a, and adds its pixel values channel by channel to the background (3.31b), causing the pure black pixels to disappear completely (3.31c).

Screen mode has an influence similar to Add mode's, but via a slightly different formula. The pixel values are inverted, multiplied together, and the result is inverted:

Note

Because Screen depends on no values being greater than 1, do not use it in a floating-point image pipeline that includes the possibility of pixel values over 1, such as eLin (Chapter 11).

Note that with this formula, fully white pixels stay white, fully black pixels stay black, but a mid-range pixel (0.5) takes on a brighter value (0.75), just not as bright as would be with Add (1).

You use Screen much like Add. Screen is most useful in situations where Add would blow out the highlights too much—glints, flares, glow passes, and so on (Figure 3.32).

Figure 3.32. Using Screen on an element brings it in more softly; notice there's less brightness in the areas of overlapping fire. A similar effect could be achieved using Add plus a Levels correction to the foreground to reduce its Output White value.

Multiply is another mode that is as simple as it sounds; it uses the formula

This would seem to make the values much higher until you recollect that we are dealing only with values between 0 and 1. Multiplying them together, therefore, actually has the effect of reducing midrange pixels and darkening an image overall, although pixels that are full white in both images remain full white.

In optical compositing (the way this was all done before we had computers) the equivalent of multiply was layering two images, one over the other, combining their densities such that the dark areas of one image held out light from the other image.

Multiply literally has the inverse effect of Screen mode, darkening the midrange values of one image with another. It is useful in cases where you have dark tones that you want to include in an image without replacing the lighter tones as much, for example to layer in texture, shadow, or dark fog to a background (Figure 3.33).

Figure 3.33. Dark smoke (actually a grayscale fractal noise pattern with no alpha channel) is multiplied over the background, darkening the areas that are dark in either the foreground or background further.

Overlay uses the bottom layer to determine whether to screen or multiply. If the top layer is lighter than 50%, it is screened. Below 50%, it is multiplied. Hard Light does the same, only using the top layer to determine whether to screen or multiply, so the two are inverse effects. Reversing layer order and swapping Overlay for Hard Light yields an identical result.

These modes, along with Linear and Vivid Light, can be most useful for combining a layer that is predominantly color with another layer that is predominantly luminance, or contrast detail (Figure 3.34). This is how you create textures; for example, a lot of the lava texturing in the Level 4 sequence of Spy Kids 3–D was created by combining a hand-painted color heat map with moving fractal noise patterns using Hard Light.

Figure 3.34. Overlay and its inverse, Hard Light, are useful for combining color and texture. Here, an instant lava lamp texture was created using the components shown at the right: a solid with Fractal Noise applied set to Overlay mode on top of a red-to-yellow gradient.

This type of usage is fine, but don't start fishing in the various Light modes when combining images. They don't work with overbright levels (explored in Chapter 11), and this method of adjusting images is a warning sign that you haven't thought things through in terms of what you're trying to do.

Difference is a subtraction mode that inverts a pixel in the background according to how bright the foreground pixel is. There is one very specific use for Difference that is never rendered: You can line up two identical layers using this mode. When all of the pixels line up properly, layer details disappear.

The Hue, Saturation, and Brightness modes each combine the given value from the foreground layer with the other two from the background layer. Saturation applies the foreground saturation to the background hue and luminance values, Hue combines the foreground hue with the background saturation and luminance, and Brightness uses the foreground luminance in combination with the hue and saturation of the background.

Color takes both the hue and saturation of the top layer, using only the luminance from the underlying background (Figure 3.35).

Figure 3.35. Setting a deep-blue-colored solid to Color mode and overlaying it on the plate footage has the effect of tinting the colors in the image blue. Subtler uses of this mode are explored in Chapter 12, “Working with Light.”

Keep these modes in mind as a shortcut to channel operations in which you might want the color from a foreground combined with the detail (luminance) of the existing background.

The Stencil and Silhouette blending modes apply transparency information to all of the layers below them in the composition. The Stencil modes use the light pixels, and Silhouette the dark pixels, of either the Alpha or Luminance values to determine the areas that remain visible in the layers below the current layer.

You should keep these in mind for that rare occasion where they save you extra setup work.

Alpha Add and Luminescent Premultiply are special-case blending modes that affect semi-transparent edge pixels only.

Have you ever tried matting a layer with an alpha channel over the same layer with the alpha channel inverted? Figures 3.36a through d show the typical result: a semi-opaque line tracing the edge of the alpha, where the pixels remain semi-transparent because they are blended together in the same manner as semi-opaque layers. Alpha Add adds the actual values of the alpha pixels without compensating for the opacity effect, so two 50% opaque pixels become 100% opaque.

Figure 3.36a through d. (left to right) Laying a matted object over a background with the inverse matte seems as though it should result in a fully opaque image (3.36a, b). Instead, edge pixels become semi-transparent (3.36c). Alpha Add does just what the title implies, adding the alpha values together so that inverse mattes total up to 100% throughout the image (3.36d).

Why would you combine a layer with itself, inverting the alpha? You probably wouldn't. But you might combine two layers with overlapping transparency that should work this way—for example, two parts of the same layer.

Luminescent Premultiply is an alternative method of removing premultiplication from source footage, retaining bright values that would otherwise be clipped. Premultiplication over black causes all semi-transparent pixels to become darker; removing the black values while adding transparency, which is what removing premultiplication does, can cause them to appear dimmer than they should. This is useful for bright overlaid elements that come in with alpha channels and premultiplication, such as flares and explosions; you can set the footage to interpret as Straight on import and use this mode instead of Normal to overlay the element. It's another trick to keep in your arsenal should you ever see the symptoms (dim translucent elements) for which it is recommended.

Track mattes are not the only way to apply alpha or luminance data to the transparency of a layer, but they're the clearest and most straightforward way. They help you out of a lot of jams in which creating a selection via other means would be inconvenient or in which mask or matte features constrain you.

For example, it's not possible to track a mask in After Effects. But it is possible to apply the mask to a track matte instead, and then to track that layer (instead of the mask itself). Chapter 8, “Effective Motion Tracking,” discusses this in detail.

In Chapter 6 you will learn of the many ways to pull a key, including hi-con mattes. Some of these methods apply their results directly to the alpha channel of the target layer, but others cannot. Additionally, specific operations such as blue-screen keying can change the color of the source (automatically removing blue spill from the foreground); applying the key via a track matte to a duplicate clean source instead is an effective workaround if you want the matte but not the despill.

Unlike with parented layers (described in Chapter 2, “The Timeline”), selecting a track matte layer does not lock it to the current layer. In fact, setting a track matte means that no matter which layer is next higher in the stack, that layer is the source of the matte. Thus, accidentally moving a layer in between a layer and its track matte can cause easily solvable but, nonetheless, disastrous results.

When you duplicate a layer with a track matte activated (Ctrl+D/Cmd+D), After Effects automatically duplicates it above the matte layer or two layers above the current layer. If you duplicate the matte layer at the same time, the duplicate will also move up two layers, so that all layers preserve their proper track mattes (Figure 3.38). That's good. What's bad is if you forget to duplicate both layers, because the track matte remains active in the duplicate layer, even if it has been duplicated to the top of the stack, and you can't see that it's active.

Figure 3.38. Selecting and duplicating the layers from Figure 3.37 creates two new layers that leapfrog above the previous layers to maintain the proper color/matte relationship in the source and duplicate layers.

If the layer to which the track matte is applied already has an alpha channel, then the new selection area created by the track matte will be opaque only in the areas that are opaque in both mattes. So applying a track matte in this situation is like making a subselection of the current selection.