Images for Video: Essentials

Here are the essential things to know about when creating and editing images for video.

Framing the Image

  • All video is landscape in orientation, which means it is wider than it is high.
  • If you want the whole image to fit the video frame, take the image in landscape, not portrait.
  • If you do take a portrait image, you can either crop it to make it fit the video frame, or animate its scale and/or position within the frame to simulate a video camera zooming and panning over the image. This technique was pioneered by Ken Burns, and iMovie includes the Ken Burns Effect to allow you to easily do this. In Final Cut Pro and After Effects, you must manually set keyframes to animate the image.

Image Resolution

  • Digital still cameras take high resolution images for printing, e.g. 3000x2000 pixels.
  • Video cameras take low resolution frames for playing back on a screen (computer/TV/projector), e.g. 768x576 pixels.
  • Therefore you do not need to use the camera's highest resolution/quality setting when taking images for video, although it can help to make them look good.
  • If you do take high resolution images, it is best to resize them before importing them into Final Cut Pro or After Effects. See Batch Resize for instructions on how to do this.
  • If doing a stop frame animation, it is better to set the quality on the camera so you get a frame size in pixels which is equal or one step larger than the video frame size you will be using. Then you'll have space on the card for all the images you'll need to take for a smooth animation.

Allowing for TV Overscan

  • All TVs overscan the video, which means that around 10% from the edge of the video frame is cropped off.
  • Projectors normally display the same area of the video frame that would be visible on a TV.
  • Web video has no overscan, so the whole video frame is visible.
  • Most editing software can display Action and Title Safe Margins (10% and 20%) to indicate the part of the frame that will be cropped off and also where best to position titles, as some TVs crop more than 10%.
  • If creating video for TV that includes still images, you may want to resize the images to 90% of the video frame and add a letterbox/pillarbox (black border) to the image, so that none of the image is cropped off by the overscan.

Matching Frame Aspect Ratios

  • See What is Frame Aspect Ratio?
  • Many compact digital cameras take 4:3 images. Some also take 16:9.
  • Many Digital SLRs take 3:2 images.
  • Video is either 4:3 or 16:9.
  • To make an image fit a video frame when they have different aspect ratios requires the image to be cropped to the correct aspect ratio, or to accept some letterboxing or pillarboxing. See Mixing Frame Aspect Ratios
Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License