Skip to content
  • There are no suggestions because the search field is empty.

Filming Best Practices

Learn how to film your content for the best possible result.

 

💡 Footage can range from very simple to extremely complex. The more dynamic the content to higher the risk for potential artifacts. Following this guide will help mitigate such artifacts 😊

 Filming best practices:
  1. Capture a wide diversity of mouth shapes
    • This is typically already captured if you have 30 second to 1min of footage of them speaking.
    • Why is this important? - LipDub studies the visemes of the actor(s) in the original video and the training data. If there isn't any lip movement in the video or the additional training footage then LipDub won't perform as well. Watch this brief 3min video to for an exmaple of this (1:41 timestamp)
    • Typically there is diminishing returns after 5min of video of the actor speaking, so there is no need to upload more than 5min.
  2. Consistent lighting throughout the video
  3. Try to avoid occlusions of the face area & extreme angles if possible
 
If you are filming extra footage on set and want best practices for filming additional training footage, please read this: How to film Training Footage.
 

For more info on what to avoid in your footage if its possible, please check out this article: