search

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

Trainning loss around 0.08? #127

Open YuTingLiu opened 5 years ago

YuTingLiu commented 5 years ago

Thanks for share this unbelivable projects. We are family// Since I decided to train my own model, uv map generating has been done, data comes form 300W-LP, which contains 122450 samples.

YuTingLiu commented 5 years ago

Solved and closed, thanks.

YuTingLiu commented 5 years ago

image This is result after 3k iters, kinds of scary. Anyway, the .obj file predicted by pre-trainned model all the same.

HBLucasKo commented 5 years ago

@YuTingLiu Hi!! Could you tell me how you solved the issue??

qiannvMao commented 5 years ago

@YuTingLiu Hi,as a newer to this, I failed to achieve the training code, could you please provide training code to me? I hope I can get your help, thank you very much. My email: mqn20170910@163.com

michaelhuang74 commented 5 years ago

@YuTingLiu Hi Yuting, where did you get the training code or did you write the code by yourself? I also like to train the model using new datasets. Do you mind share your training code with me? Thanks a lot. My email is mqhuang@gmail.com

JianBin-Liu commented 3 years ago

Thanks for share this unbelivable projects. We are family// Since I decided to train my own model, uv map generating has been done, data comes form 300W-LP, which contains 122450 samples.

  • What I am not sure is the trainning loss always around 0.08, output as follow: FireShot Capture 046 - TensorBoard - 192 168 5 109 Can someone tell me how this happen? my trainning loss always around 2.00, did I need to train consistently?
JianBin-Liu commented 3 years ago

Thanks for share this unbelivable projects. We are family// Since I decided to train my own model, uv map generating has been done, data comes form 300W-LP, which contains 122450 samples.

  • What I am not sure is the trainning loss always around 0.08, output as follow: FireShot Capture 046 - TensorBoard - 192 168 5 109 Can someone tell me how this happen? my trainning loss always around 2.00, did I need to train consistently?