Mr DeepFakes Forums
  • New and improved dark forum theme!
  • Guests can now comment on videos on the tube.
Total Likes Received: 2 (0.08 per day | 0.15 percent of total 1304)
(Find All Threads Liked ForFind All Posts Liked For) Total Likes Given: 7 (0.29 per day | 0.54 percent of total 1305)
(Find All Liked ThreadsFind All Liked Posts)


Registration Date: 06-01-2019
Date of Birth: Not Specified
Local Time: 06-25-2019 at 03:33 PM

VirginBoI's Most Liked Post
Post Subject Numbers of Likes
RE: How Does Training Work? 2
Thread Subject Forum Name
How Does Training Work? Questions
Post Message
(06-22-2019, 09:32 AM)LCC Wrote: You are not allowed to view links. Register or Login to view.For each destination picture, does the trainer only use source pictures with a similar face angle? 

Or is a free-for-all, with any image being chosen?

As per what I have observed  if there are a face similar in angle in src and dst then it is pseudo independent of that. Basically it chooses pictures of both faces  and  learns to draw it simultaneously  . later it just switches to creating a src face where it should create dst .

Kind of you learn two languages (src and dst facesets) and all alphabets of them  ( here faces) and later just translate what written in second language sentence (dst faces) into your learned first language sentence making use of its alphabets which closely resembles to the meaning to what written in second .

So yes its pseudo independent as if there are no faces of similar angle in src like in dst then It will simply not know how to translate that dst face specific angle into prediction. Like there is no alphabet equivalent of that second language (dst) alphabet in first

VirginBoI's Forum Info
Last Visit:
10 hours ago
Total Posts:
63 (2.62 posts per day | 0.87 percent of total posts) Find All Posts
Total Threads:
16 (0.67 threads per day | 1 percent of total threads) Find All Threads
Time Spent Online:
3 Days, 21 Minutes
Given: 7 | Recieved: 2
Members Referred:
VirginBoI's Contact Details