Training the model

[Tinaa23]

Thank you for providing your code.
In the train.py file, driving.png and driving_r.png are used.
Can you please clarify what is the difference between these two? and Is this model limited to training only two frames at a time?

[KenChen701]

I think driving.png and driving_r.png are used as a kind of validation. driving.png is for cross-identity reenactment while driving_r.png is for same-identity reconstruction.

[Tinaa23]

I think driving.png and driving_r.png are used as a kind of validation. driving.png is for cross-identity reenactment while driving_r.png is for same-identity reconstruction.

Thank you. It makes sense now.