Page 1 of 1

Need more details

Posted: Wed Dec 02, 2020 2:52 am
by gseinstein

After successfully completing 100k iteration I generated a video even though it replaces the face perfectly it is quite blurry (no details). Can more details be achieved by training for more time ??


Re: Need more details

Posted: Wed Dec 02, 2020 4:18 am
by abigflea

Yes , more time.
Also there are more variables to making a reasonable/high quality swap.

Number of training images on both sides.
Image quality of training data.
Which model.