yfeng95 / PRNet

Joint 3D Face Reconstruction and Dense Alignment with Position Map Regression Network (ECCV 2018)
http://openaccess.thecvf.com/content_ECCV_2018/papers/Yao_Feng_Joint_3D_Face_ECCV_2018_paper.pdf
MIT License
4.95k stars 947 forks source link

a small error in train.py #114

Open zhuixunforever opened 5 years ago

zhuixunforever commented 5 years ago

https://github.com/jnulzl/PRNet/blob/bbdc3c95266aa4ce6379cd343121616244964872/train.py#L39 this should be 'label', instead of 'img', like as: label_array = np.array(label,dtype=np.float32)

by the way, thanks for your train code!

destiny311 commented 5 years ago

label_array = np.array(label,dtype=np.float32) can you tell me the datasets where i can get:?

LucienXian commented 5 years ago

@destiny311 I have met the same problem, have you solved it? What is "--train_data_file“

zhuixunforever commented 5 years ago

@destiny311 @LucienXian I get the dataset by search "300W-LP" in Google, and you can find the download link of 300W-LP in the web http://www.cbsr.ia.ac.cn/users/xiangyuzhu/projects/3DDFA/main.htm.

I generate groundtruth position map just using generate_posmap_300WLP