LizhenWangT/FaceVerse

Some questions about this project

Opened this issue · 2 comments

nice job!I have a few questions, hope to get your reply
q1.Will the training part and the CUDA accelerated rendering version be announced soon?
q2.For example, how to calculate 52blendshapes for your V2 version of the design process?
q3.Do you have an idea for an end-to-end training and inference solution? Can you provide a case or open source project to learn from?

Thank you!

  1. Sorry, we do not have plan to release the CUDA code recently.
  2. Maybe you can refer to this issue.
  3. A new ECCV 2022 paper done by Microsoft provides a quite stable result using a dense landmark prediction network. We'd like to follow their work recently.

Thanks for your answer, here is an engineering question to ask you: is the output blendshape communicating to UE or using arkit, and if it is Arkit, how does this connect?