XingangPan / GAN2Shape

Code for GAN2Shape (ICLR2021 oral)
https://arxiv.org/abs/2011.00844
MIT License
571 stars 104 forks source link

How to save images for synface pre-training #14

Closed junxuanlee closed 3 years ago

junxuanlee commented 3 years ago

Hi, may I know how can I save the images in synface pre-training? As what I have seen, the save_results function in model.py is coded in a way to receive one image path and generate image accordingly. But in the synface pre-training, a list of image paths is presented to the save_results function instead, I suspect it may be caused by the joint_train = true.

Is there a way that your code for the save_results function can be changed to save images from joint training or the function was never intended to save images from joint training? Does that mean I need to write my own code to generate images from synface pre-training?

image

Error encountered: image

XingangPan commented 3 years ago

@jun1byte Hi, you may check out the visualize_results function in model.py, which saves some images in tensorboard.