Take two inputs for one output
lucydjo opened this issue · comments
Hello,
I would like to have your opinion on the following feasibility study:
Is it possible to modify pix2pix to include :
Input 1: Image to be modified
input 2 : Vector image "prompt
Output : The image modified with the instructions of the pormpt image
The idea is to be able to modify the shape of the lips while retaining the original appearance. ( skin color, beard ).
I've created a dataset with 5000 images of lips/mouths and their position/lip shape, teeth...
Like this :
What do you think?
Can't wait to hear what you think!
Thx