I have an image of a face, facial landmark positions of it:
And another image with the landmarks at other positions:
I need to stretch the first image by "dragging" the landmarks to the new positions which are in the second image, in Python (I'm trying to create a very basic deepfake).
How can I implement this? (I have the exact position of the landmarks)
I searched a lot in Google but haven't found any result, so I hoped someone here can help me.
Asked
Active
Viewed 128 times
0
Yuval.R
- 1,182
- 4
- 15
-
I do not think this is doable without deep learning. – tino Feb 25 '21 at 09:47
-
Really? There is no simple way to do this? – Yuval.R Feb 25 '21 at 09:56
-
I don't think so. It is not a simple composition of translation and rotation: you are actually warping the image. Absolutely a non-trivial process. – tino Feb 25 '21 at 10:00
-
In that case do you have any idea where to look for a solution? – Yuval.R Feb 25 '21 at 10:08