Page 1 of 1

Different alignment files?

Posted: Thu Jun 03, 2021 6:34 pm
by jenkaslay133

Hi there, i have extracted the faces from several different videos and also some other pictures, they all have their own and individual alignment files, does it cause any trouble ? if so is there any way to extract faces from different videos all at once and is classified in the same alignment file.

furthermore, i have receive an error while trying to clean the alignment files

the error is as below ...

Code: Select all

Traceback (most recent call last):
  File "C:\Users\User\faceswap\lib\cli\launcher.py", line 182, in execute_script
    process.process()
  File "C:\Users\User\faceswap\tools\alignments\alignments.py", line 40, in process
    job.process()
  File "C:\Users\User\faceswap\tools\alignments\jobs.py", line 678, in process
    self._update_png_headers()
  File "C:\Users\User\faceswap\tools\alignments\jobs.py", line 712, in _update_png_headers
    face.from_alignment(self._alignments.get_faces_in_frame(frame)[new_index])
IndexError: list index out of range

as index out of range , i am guessing the index which for the number for each images have to be exact as the alignment file,
but i might be wrong tho, can anybody help me


Re: Different alignment files?

Posted: Sat Jun 05, 2021 9:04 am
by torzdf

I have extracted the faces from several different videos and also some other pictures, they all have their own and individual alignment files, does it cause any trouble ?

For different sources to have different alignments files? No. This is how it should work. Not sure what trouble you would expect here?

if so is there any way to extract faces from different videos all at once and is classified in the same alignment file.

No. A single source (be it a folder of images or a video) = A single alignments file

As to your error message. I can confirm that I use remove-faces often and have never hit this issue. It might be better for you to go through the entire workflow steps that you are undertaking that brought you to this point, because most likely you are making a mistake somewhere.