How to obtain the results in the teaser?

papulke / face-of-art

Code for "The Face of Art: Landmark Detection and Geometric Style in Portraits"

MIT License

261 stars 31 forks source link

Open Honlan opened 4 years ago

Honlan commented 4 years ago

hi~ first of all, thanks a lot for your awesome work and open-source codes.

It seems that predict_landmarks.py only produces the landmarks.

How to obtain the transfer results in the teaser, namely the rendered faces given the reference art works?

JaeDukSeo commented 4 years ago

really would love to know this as well!

Nauhcnay commented 3 years ago

They provided a function called "create_img_with_landmarks " in logging_functions.py, which works well for me. But these need some adaptation:

remove parameter "image_size" and "scale" and corresponding codes in that function
open the input image as a NumPy array, convert "pred_lms" to NumPy array and pass them to the adapted function
save the returned result

But it still will be great if there would have a better way to visualize the prediction results as the teaser does :P