Hi,
I am new to all of this. So, apologies if this question is totally stupid.
I would like to shoot a video altering the face (obviously). However, rather than doing a straight swap, I would like to just change the features -- for example (and it's my real use cases) make eyes bigger, and nose smaller. It will still look like the original face (more or less) but not quite.
The only way I can think of doing this is by getting a sample of photos of the original person, and then train the system with the same sample photoshopped. However, this would be difficult at best since each editing will be done by hand.
Is this something that is already been toyed with?
Thanks.