Converted vid only swaps 1 FPS, so I need to re train

Want to understand the training process better? Got tips for which model to use and when? This is the place for you


Forum rules

Read the FAQs and search the forum before posting a new topic.

This forum is for discussing tips and understanding the process involved with Training a Faceswap model.

If you have found a bug are having issues with the Training process not working, then you should post in the Training Support forum.

Please mark any answers that fixed your problems so others can find the solutions.

Locked
User avatar
Yosarian9
Posts: 1
Joined: Thu Mar 23, 2023 8:19 pm

Converted vid only swaps 1 FPS, so I need to re train

Post by Yosarian9 »

I extracted face A at every 25 frames(because the guide says that too many faces isn't better and 97k was the whole video) and then trained for 170k iterations with face b. I'm not totally done yet as it's still blurry a little but i wanted to convert just as a test. But when I did the faces are only swapped once for every 25 frames. I'm now extracting face A at every frame. Do I need to retrain again or can I just clean up and use the new alignment file?

User avatar
torzdf
Posts: 2636
Joined: Fri Jul 12, 2019 12:53 am
Answers: 156
Has thanked: 128 times
Been thanked: 614 times

Re: Converted vid only swaps 1 FPS, so I need to re train

Post by torzdf »

No need to retrain. You just need to generate an alignments file for the target conversion video. See here for more detail: viewtopic.php?t=27#align

My word is final

Locked