You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
To run demo1 on an outside image you need to align the image and generate LFW attributes. Face alignment code is out there. I'm not aware of an LFW attribute classifier implementation, but you could train one using the LFW data.
It might be easier to begin with demo2 and replace deepmodels.facemodel_attributes() with your own AttributeClassifier model which predicts LFW attributes and selects from the LFW image database.
Hi,
I was wondering if it would be possible to run demo1 on an image outside of the LFW dataset.
I understand I would have to generate the attributes entry in the format of the lfw_attributes.txt file.
Do you know of an open implementation i cold use to generate the LFW-like attributes for an arbitrary image?
Thank you
The text was updated successfully, but these errors were encountered: