anhttran / 3dmm_cnn

Regressing Robust and Discriminative 3D Morphable Models with a very Deep Neural Network
Other
857 stars 254 forks source link

solver.prototxt for training? #12

Closed dingtiandu closed 6 years ago

dingtiandu commented 7 years ago

Can you share your solver.prototxt for training?Thanks.

wanggongziZeo commented 6 years ago

Do you have the file now?

iacopomasi commented 6 years ago

Hi, we are not planning to release the training configuration file now but we might do it later on. You can use the deploy.prototxt and use it for training: just adjust the Batch Normalization configuration and add L2 loss function. More advanced loss functions can be implemented using caffe layers.

Thank you.