Creating cutouts

At present there is no custom cut-out feature, BUT if you have some useable images, are willing to do a little work and to share with others, I can add yours to the main library.

Each cutout consists of two files - the image file, and a definition file.


The image file is a transparent PNG file, sized to fit within the VidProv stage on a typical screen.

  • format: PNG
  • preferred height: 989px
  • preferred width: 1600px - 1920px

To create a nice looking cutout, it is recommended to use Photoshop or similar and manually cut the holes in the original image.

  1. Open the source image in Photoshop (or similar), and if necessary, convert it to a layer.
  2. Crop any unwanted parts of the picture, aiming for a nicely framed horizontal rectangle.
  3. Resize the image to have a height of 989 pixels.
  4. Use the lasso tool, set Feather to 4px, and cut out face-shaped holes in the image.
  5. Go to File > Export > Quick Export as PNG, to save the file. Save the file with a meaningful name. E.g. if a picture of 2 doctors, then "doctors.png".


  • For human or animal faces, it's best to leave as much of the original as possible, so only cut from just above the eyebrows to just below the mouth.
  • 4px seems like a nice value for feathering to provide a soft blended edge, but in some cases you may want more or less blending.
  • You can experiment a bit in Photoshop by adding a face image as a layer below the main image. Just be sure to hide those layers before exporting.


Next, we need information about the file and where the video should appear. The file will look something like below. Use this as a template for your file.

  "width": 1800, "height": 989,
  "hole": [
    {"x": 568, "y": 291, "w": 255, "h": 373},
    {"x": 1115, "y": 299, "w": 255, "h": 387}

We can find the information we need using Photoshop.

  1. Open a text editor, and copy the example file into it. Include everything from the opening "{" to the closing "}".
  2. If you've closed the image file, reopen it in Photoshop.
  3. Go to Image > Image Size and note the width and height of the image. Enter these into your file.
  4. Make sure the information tab is showing. This will be used to get information needed.
  5. For each hole...
    1. Move the cursor to centre it within the cut hole.
    2. Note the x and y coordinates.
    3. With the rectangular marquee tool, select an area around the hole, that is about 3 times the height of the hole itself, and at least wide enough to completely capture the hole.
    4. Note the W (width) and H (height) values.
    5. Enter the x, y, w, and h values in the file.
  6. Save your text file with the same name as the image but with a ".json" extension. E.g. "doctors.json".

Send files

Send both files to Jonathan.

Hopefully, this will become a little more built-in in time.