search

LizhenWangT / StyleAvatar

Code of SIGGRAPH 2023 Conference paper: StyleAvatar: Real-time Photo-realistic Portrait Avatar from a Single Video
BSD 2-Clause "Simplified" License
407 stars 44 forks source link

Question about demo video #25

Open Jerry-Master opened 1 year ago

Jerry-Master commented 1 year ago

Sorry for bothering you but I saw in the demo video that you manage to reenact other faces like Obama's. How did you do this? Because as specified in the article, the generated model can only produce expressions close to those available in the dataset you train styleavatar with. Did you find such images for Obama, or is there a way to achieve same result training differently?

Inferencer commented 1 year ago

Sorry for bothering you but I saw in the demo video that you manage to reenact other faces like Obama's. How did you do this? Because as specified in the article, the generated model can only produce expressions close to those available in the dataset you train styleavatar with. Did you find such images for Obama, or is there a way to achieve same result training differently?

I'm just about to start learning style avatar but I would imagine they used video frames to train rather than searching for individual images