search

tzt101 / MichiGAN

MichiGAN: Multi-Input-Conditioned Hair Image Generation for Portrait Editing (SIGGRAPH 2020)
MIT License
294 stars 39 forks source link

How to reshape hair with inference.py script? Description of preprocessing pipeline. #6

Open TsykunovDmitriy opened 4 years ago

TsykunovDmitriy commented 4 years ago

Thank you for your work!

I run the inference script many times with different flags. Could you tell me how to achieve hair reshaping? Could you describe the preprocessing pipeline and input data format? I wasted a lot of time trying to run the inference not on a FFHQ dataset.

tzt101 commented 4 years ago

Hi, the inference can not change the shape of hair, if you want to do it, please run demo.

I think you might be wondering about the orientation input. The size of orientation map is H*W, ranging from 0 to 255. You can refer to the codes from ./data to find more details.

shionhonda commented 3 years ago

@tzt101 What makes the difference in functionality between demo.py and inference.py? I dig into the code but could not figure out. Why can demo.py change hair shape and inference.py cannot?

(demo.py shows just a blank window for me.)