Mr DeepFakes Forums
  • New and improved dark forum theme!
  • Guests can now comment on videos on the tube.
   
iperovAnnouncing improved FANSeg model for converter.
#1
del    .
#2
would be useful if the video actually showcased the tech...
#3
Just so i understand, after editing ( (5.4) data_dst mask editor) just send the contents of /data_dst/aligned_confirmed . That's it, no other files needed?
#4
Looking forward to giving this a try once it is implemented.
#5
Let's make blowjobs great again guys! I'll mask some of my scenes and email soon.
#6
(04-13-2019, 05:27 AM)mondomonger Wrote: You are not allowed to view links. Register or Login to view.Looking forward to giving this a try once it is implemented.

It's already in the current build
#7
If it's in the current build, has anyone given a go at a blowjob clip? I'm not in a position to be able to work on this until next week, but will certainly give it a go.
#8
its not in the current build.
why ppl did not read change log?
what wrong with word "Announcing" ?

[Image: 226679318_81a547b3ee233bffa1417205741fb2f2_800.gif]
#9
nevermind, I started to train fanseg for production in google colab.
#10
@iperov I have couple of questions.
- So to use fanseg I have choose that option during convert. Which option do you recommend?
- If model learned mask shouldn't I use it? Otherwise I might disable learning mask during training.
- if we send you bj pictures would you train fanseg more? If so what kind of pictures do you need?

Forum Jump:

Users browsing this thread: 1 Guest(s)