Andrew Somers
Forum Replies Created
-
In After Effects layers themselves are always flat. However, you can add effects to warp the video inside the layer.
Add an effect to the layer such as MESH WARP, which is under EFFECTS > DISTORT.
In the reference clip you presented above, there is also tracking involved, and you may want to try Mocha AE for this task since it works well for placing 2D video into a 3D space, as well as creating tracking masks.
-
If you have an open PCIe slot, you can get a BlackMagic Intensity card.
Otherwise, just make a composition viewer, and drag it to your program monitor, and maximize the window in that monitor. You have independent control of “Display color management” and other features for each viewer window via the bottom row of buttons.
-
Smoke or steam when backlit and bright, such that it would be served well by additive transfer, or additive transfer *combined* with multiply. You need the MULT layer to reduce brightness and transparency.
Also, bright elements are especially good for black screen – sparks for instance add especially well in 32-bit linear space. In additive xfer these kinds of elements are essentially self-keying.
I’m setting up to do this right now to augment a shot with a cutting torch/sparks. (An acetylene torch could not be used too close to the actors on the day for safety.)
So, I’m going to shoot a real torch element against duvetyne, cutting sheet metal that has been painted flat black. The sheet metal will be oriented close to the orientation of the steel being cut in the actual scene, and shot as a static lockdown (the nature of the shot should allow cheating perspective, even though the original camera was not moving as nodal). Then tracked and comped into the scene using ADD.
The *really cool* thing about working in 32-bit linear is the ability to “ADD” and also combine ADD and MULTIPLY to really dial things “into” a shot.
Still, I don’t mean in anyway to imply that black-screen replaces green/blue/red – it is a separate tool, with its own quirks and “usefulness curve”. But the next time you need to comp backlit steam, or bright things like sparks and flames, or solid objects with low contrast I think you’ll be happily surprised at how quickly you can dial in a comp.
Note also the “gamma/blur” technique I mention above can also be used in color key situations. Instead of doing the actual comping inside the keying plug-in (keylight, primatte, whatever), just use the plug in to create the matte(s), and then use gamma/blur, and COLOR/ADD/MULT transfer modes (and using precomps to pre-mult etc) for bringing it all in together. While this may sound more complicated (okay, it *is* more complicated), the level of control you gain is often worth it IMO.
Side note: IMO, Ultimatte Advantedge was the only plug-in that really worked well as an “all in one” comping solution, but sadly it is not longer supported, and even if it was, it was *DOG* slow to render. Ultimatte is now focused only on their hardware solutions.
-
You can do some of those effect using expressions/masks etc.
But by far the best particle systems for After Effects are Trapcode Particular, and also related, Trapcode Form and 3D stroke. You can get them all in the suite:
https://www.redgiantsoftware.com/products/all/trapcode-suite/
And here is a collection of tutorials on Trapcode, with lots of examples similar to what you were asking about:
https://www.rnel.net/after_effects-tutorials/the-best-trapcode-tutorials-on-the-internet
-
It’s normal that you cannot see the model in the Top, Left, Right camera. Since 3D Invigorator is a plug-in that is applied to a 2D Layer, the 3D is only “projected” to the Comp Camera.
Interestingly, Trapcode Particular is a 3D plugin applied to a 2D layer – but you *can* see it in the top/right/front etc. views.
I have to tell you it is very handy for positioning. Not sure what Trapcode is doing to bring this functionality, but I’d imagine you could do the same for invigorator.
-
To answer the question you asked, use a matte choker, such as “Simple Choker”, plus blur, and if needed, masks. Mocha can be very helpful here.
In CS5 and later, look at using Rotobrushes and refine matte – though personally I find these new tools frustrating to use:
https://help.adobe.com/en_US/aftereffects/cs/using/WS3bf812c123007fb8513559df126b537c840-8000.html
To answer the question you didn’t ask – which I am assuming is “how to get a good comp with black screen”, I’ll continue – but first, can you post a still image of the clip you are trying to comp?
Black screen can be the ideal background for some elements intended for compositing – for instance, transparent elements such as grey smoke, water, or steam can be great against black screen as you can often composite them by simply using the additive transfer mode (when working in linear 32-bit space, or “screen” in non-linear spaces).
Black screen can also bee useful for solid objects that have sufficient lighting and limited dark areas. But a typical mistake here is to try and use the luma key plugin, or to just set a layer with levels clipped and use it to luma key.
INSTEAD, for solid objects try the hold-out matte approach.
STEPS:
1) Work in 32 bit LINEAR color space.2) Create a comp with your background plate.
3) On top of your background plate, place TWO layers of the black screen element.
4) On the middle black screen layer, place the following effects, in this order:
Top: Levels Individual Controls – name it “invert and matte”
Middle: Blur – name it “matte soften”
Bottom: Levels – name it “matte choke”5) Adjust as so:
* Set the top Levels, named “invert/matte”, so that the black-screen area is at 1.0 value (32 bit mode white). Adjust so that the brightest area of the subject is 0.0 (black). Set the output BLACK level to 1.0, and the output white level to 0.0. Set “clip” to ON for both black and white. Using the pickwhip, connect the INPUT while level to the input black level, then adjust the input black level until the black screen is clipped white, and you are left with a black solid for the subject.
* *Optionally* drop in a TINT effect to make the clip greyscale *or* use the pickwhip to connect the input white/black for each of the RGB channels, and clip each color independently.
* Set the Blur effect to 10 pixels initially..
* Leave the bottom Levels alone for now.
6) Set this layer to MULTIPLY transfer mode.This will put a solid black area on the background plate.
7) Now, with the TOP black-screen layer, set to ADD transfer mode. If necessary, add a garbage mask, and a levels control to match the background plate.
8) Now the magic touch – Adjust the GAMMA (only) on the bottom Levels (matte choke) on the middle black-screen layer to get a clean edge. Gamma will work with BLUR, to adjust the spread/choke of the cutout matte. Adjust the blur pixels in association with gamma, to get the matte edge that you like.
SInce you are ADDing the topmost layer, you can adjust edge transparency without clipping the edge of the foreground element.
If your subject is a combination of solid and transparent objects, you will need to add a mask to the middle black-screen layer so that the areas of transparency are clipped to white.
To adjust the density of the transparent portions, duplicate the middle layer, and use the same mask to set all but the transparency region to white. Adjusting the opacity of this layer will adjust the density of the transparency.
It should be obvious that black-screen will not work well when the the subject has dark areas that are within clipping distance of the black screen area. Also, this method tends to break down with 8 bit images. Ideally, you want to be capturing at least in 10 bit LOG, or 12 or 16 bit linear (RAW), with the black-screen at least a stop below the darkest values in the subject.
-
You normally do NOT need tracking markers, IF there are enough high contrast stationary points in your set. In the video you showed as an example, there is more than enough detail in the bridges and buildings .
I have seen all too often in footage, well meaning people in production putting tracking marks all of the the place where not needed – inly to create a HUGE problem in post in painting them out.
You need specific markers:
1) On green/blue screen backgrounds, when the camera is moving – you need tracking marks to define the screen.
2) On moving subjects *if* those subjects are going to be replaced with CGI. (i.e. face dots for face replacement, and body dots for body mocap/replacement).
3) On things that need to be tracked that have no high contrast and stable geometry that is trackable, such as smooth monotone surfaces.
One place people often put tracking markers where they are typically NOT needed is on monitor faces. This is a HUGE pet peeve of mine. If you can see at least three of the monitor corners in the footage, then you do NOT NEED MARKERS. You mainly need markers on a monitor face if you cannot see the monitor’s frame. in fact, you cane use a green image in the monitor with rectangles of same saturation but different luminance, and have the best for tracking and for keying.
-
Are all of your comps at the exact same frame rate?
-
If, on the other hand, all you want to do is put the actor into an “ironman” suit that folly encompasses his body, then you can just shoot the actor in a monotone skin tight suit covered with high contrast dots, and shoot him on your actual set. When you shoot him, you need to use your main camera for the plate, but you must *also* cover him with at least two additional “witness” cameras (in genlock/frame sync).
Then use something like PFtrack and extract the motion.
Built the “suit” in a 3D app, using the motion/mesh from PFTrack to animate, render and then composite over the actor.
A “green screen” suit is not needed in this case – black with highly reflective dots is probably preferable.
-
First, to echo what Michael was saying, there is a huge misconception that if you put an actor in a green or blue suit that you can then press a magic keying button and make him disappear – nothing could be farther from the truth.
It *is* useful to have the actor in a monotone blue or green (or red, depending) suit to make auto-tracking masks easier, IF auto-tracked masks will be needed.
What is more important is the CLEAN PLATE that is the scene behind the actor – how are you going to capture that? Either your camera is locked down, or you are using a motion control rig with repeatable moves. OR your are going to shoot only the PLATE, and then get the motion from a stage.
And what level of motion capture do you need? Do you need it to drive a fully articulated “robot” 3D character?
Which method is best is hard to say, since you have not really told us anything about the shot.
So I will make some assumptions – You don’t want to see the actor at all, only the “”3D robot”, against the background. With this assumption, here is *one* possible workflow:
Shoot the plate for the shot where you want the action to occur. You should have the actor there for some rehearsal takes to nail down the camera move you want, but shoot it with the actor NOT in the shot at all. This could he hand held, on a dolly, or however you like – motion control is not needed, but a stop watch and someone calling out “key points” to match the rehearsals would be useful. (You should ideally shoot some of the rehearsals as a reference).
Take accurate measurements of some of the prominent features of the environment (exact size of door frames, etc) as this will be useful is setting scale in the 3D track pass, next.
NOW: Take your CLEAN PLATE, and do a complete 3D motion track using PFTrack, BouJou, SynthEyes, etc.
From this clean plate you can determine the “environment”.
Now, Go to a stage to shoot the motion – put the actor in a skin-tight suit, covered with very high contrast dots, and shoot him with multiple cameras at the same time (from different angles). It is important that these cameras are all “genlocked” – i.e. the shutters are all “in sync: with each other, all opening and closing at the same time.
If necessarily, add elements on the stage that match physical properties of the background clean plate (a chair, a railing to jump over, whatever).
Load this footage into a mocap program and use that motion mesh to develop your 3D objects and animation, then render than and composite it on the clean plate.
That is the *basic* run down – in practice, it usually takes a *team* of people that are specialists in each phase. If you are planning on doing this all yourself, expect the learning curve for each phase of development.