How create model from aligned images?

Home Forums DeepFaceLab Training How create model from aligned images?

  • This topic has 1 reply, 2 voices, and was last updated 2 years ago by deepfakery.
Viewing 2 posts - 1 through 2 (of 2 total)
  • Author
    Posts
  • #7852
    erikdenice
    Participant

      Could someone tell me how to do step by step to create a model for Deepfacelive (without initial video but just with aligned images). I’m new to Deepface and would like to create some models and use them in Deepfacelive. I can’t find any video tutorial which would be really interesting. Simple step by step explanations with a concrete example. If someone can explain this to me it would be really cool.

      Erikdenice

      #7868
      deepfakery
      Keymaster

        Have you seen this tutorial by Druuzil: https://www.deepfakevfx.com/tutorials/deepfacelive-tutorial/
        Its not simple but he shows all the steps.

        You need to train a model using your chosen SRC faceset and a variety of DST images, such as the generic pretrain faceset, RTM faceset, etc.
        You can speed up the entire process by starting with a pretrained model.
        Once this SRC-generic model is trained you have to export as DFM.
        After that use the DFM file in DeepFaceLive.

      Viewing 2 posts - 1 through 2 (of 2 total)
      • You must be logged in to reply to this topic.