Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Can use pytorch directly train the mtcnn? #3

Open
guanfuchen opened this issue Feb 4, 2018 · 4 comments
Open

Can use pytorch directly train the mtcnn? #3

guanfuchen opened this issue Feb 4, 2018 · 4 comments

Comments

@guanfuchen
Copy link

first thanks for your sharing code, it is good for me to use pytorch get the model test, I want to know can this model be trained directly using pytorch, instead converting the caffe model to npy.

@TropComplique
Copy link
Owner

You can't train MTCNN using this code. I haven't implemented training because it's quite burdensome.

I believe that it is quite hard to train MTCNN. And if you want to train your own face detector then consider SSH or FaceBoxes. I think they are more powerful, faster, and easier to train.

@KoapT
Copy link

KoapT commented May 30, 2018

You havent implenmented training? Where are the weights of the 3 nets from?

@minhng92
Copy link
Contributor

He converted the weights from the pretrained Caffe models. Details can be found here:
https://github.com/TropComplique/mtcnn-pytorch/blob/master/extract_weights_from_caffe_models.py

@guanfuchen
Copy link
Author

@TropComplique Thanks for your reply, I will try these two models(SSH and FaceBoxes).

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

4 participants