Skip to content
Disguise User Guide

Green Screen Workflow

  1. Set up the keyer for video input using one of the following two workflows below:
    1. Setting up keyers -
    2. Setting up keyers - Hardware keyer
  2. Set up tracking for the camera (if not static) and perform spatial calibration. This can be one of three techniques listed below.
    1. Calibrate the tracking system outside Disguise.
    2. Or use an LED screen to perform the usual spatial calibration workflow.
    3. Or take manual observations to calibrate your camera as show at the bottom of this page.
  3. Calibrate the camera’s tracker delay.
  4. Ensure the camera’s video receive delay is set to 0.
  5. If you want to use set extension, model the green screens and add them to your project.
  6. Map content to camera front and backplates as in the standard XR workflow.
  7. To achieve camera switching, add your cameras to an MR set as normal.

Setting up keyers

a. Notch

  1. In the video patch editor, add layer stack on the videoin.
  2. Create a new notch layer in the layer stack.
  3. Select chroma keyer notch block.
  4. Ensure blend mode is set to Over-Alpha.
  5. Calibrate notch keyer using the instructions at Chroma Key.

b. Hardware keyer

  1. Connect the cameras feed into the input of the hardware keyers.
  2. Connect the Key and fill feed from the keyers to d3 machines ( Mustbe connected to 2 separate video inputs).
  3. In the video in patch editor, create a video in or select the “new luma keyed input” opt
  4. Map the chroma and luma key souces.
  5. Select the luma key to Opaque white or black depending on the format coming from the hardware keyers.

Camera calibration using manual observation

  1. Accurately model a real-world object within d3 to use for calibration. A flat-screen won’t work very well as you need at least 6 points for calibration. A (cuboid) box is an easy way to do this.
  2. Open up the spatial calibration editor from the camera.
  3. Under Manual Observations, add your alignment object.
  4. Under Calibration, set your calibration scope to Registration only if you already have calibrated lens values, or Registration and focal length if not. Manual observations are unlikely to contain enough points to accurately calibrate the lens distortion parameters.
  5. Click Add manual observation. This will take a picture of the stage and capture the current tracking data. The new observation will appear under Observations.
  6. Right-click on the new observation to open up the alignment editor, similar to Omnical’s QuickAlign tool.
  7. If you do not see the wireframe of your alignment object, increase the stage render opacity to see a render of the stage, and adjust the solved pose manually to get the object into view.
  8. Left-click on the vertices of the alignment object wireframe to add alignment points, and drag these to the corresponding points in the captured image.
  9. Once you have 6 or more points aligned, click update calibration. The wireframe of the alignment object should move to align much more closely with the real object.
  10. If you are using a static camera, the calibration is completed.
  11. If the camera is not static, similar to the XR workflow, keep moving the camera and taking and aligning observations until you are happy with the calibration over the whole movement area of the camera.
  12. You can also take observations at different zoom/focus levels to calibrate the lens if required.

Using set extension

  1. Create green screen objects in d3 from Stage > MR Sets. Model the screens as accurately as possible, although it’s best to keep them slightly smaller than the real green screens to allow for small inaccuracies.
  2. Ensure your camera is spatial calibrated.
  3. Create a new MR set, add the green screens to the on-stage list, and add your camera via the override parameter or using an indirection controller.
  4. Any backplate content outside your camera’s view of the modeled green screens will be rendered as a set extension.

Green Screen Setup with Layer Stack

With the addition of the new Layer Stack feature in r20.0 it is now possible to do a quick Green Screen setup while applying a Notch Chroma Key asset to a VideoIn object and here is a quick way of how to do that.

What you need for Basic Green Screen setup

  • A Green Screen or a green cloth that is similar to a green screen that is positioned in front of the camera you will be using. In this basic setup we’ll be using the webcam of our work laptops but any camera can be used with this workflow.
  • The Notch block with the Notch keyer, which will need to be added to your d3 project’s NotchFile folder.

How to setup Green Screen with Layer Stack

  1. Launch new d3 project on any r20.0 build.
  2. Add Video input to VideoIn object.
    1. Right-click on the ‘Devices’ tab at the top of screen
    2. Click on Video Input Patch Editor.
    3. Click on the black window under VideoIn_1 and then expand the Input configuration for VideoIn_1 field to open its input configurations.
    4. Now click on Input and choose the camera feed of your choice. In case your don’t have a real camera connected to d3, you can use your laptop’s web camera.
    5. Look for names like USB2.0 HD UVC WebCam.
    6. Click on Start Preview under the VideoIn_1 to make sure the camera is working alright.
  3. Create new Layer Stack.
    1. While in the input configuration field for VideoIn_1, click on Layer Stack.
    2. Insert the name for your new Layer Stack and click OK.
  4. Add and configure the Notch layer in Layer Stack
    1. While in your Layer Stack, give a name to the layer you want to create and then from the list that comes up select a Notch layer.
    2. Then right-click on the newly created layer in your stack and press Edit.
    3. Now go to the Effect field and select the block with the Notch keyer.
    4. Open the Default tab and change the Blend Mode to Over-Alpha.
    5. Under the Image 2D tab, change the Image Texture to your VideoIn_1.
  5. Add VideoIn and Content to your Surface
    1. To your timeline add a Video layer named Camera that is mapped to Surface 1 and as content add your VideoIn_1 to it.
  6. Underneath the Camera layer add another Video layer, that’s mapped to Surface 1 as well and to it add any content you like.
  7. Observe how the content gets applied to the green screen correctly and that any real world objects in front of it appear as intended.