Davis E. King

Some things are lost when converting from batch norm to affine. Save in batch norm form when doing training. On Thu, Oct 31, 2019 at 9:17 AM Claudiu wrote: >...

You could copy the layer parameters into a network with batch norm layers and go in and manually fill in the missing batch norm parameters with something reasonable.

http://dlib.net/dnn_face_recognition_ex.cpp.html and http://dlib.net/dnn_metric_learning_on_images_ex.cpp.html explain all that. On Wed, Sep 9, 2020 at 6:06 AM ismailpolas wrote: > Does anyone build the dlib_face_recognition_resnet_model_v1.dat with > custom dataset? If Yes, Please share...

There isn't anything special about training a gender model vs any other image classifier. So the normal docs in dlib showing to to train an image classifier using a CNN...

no On Fri, May 29, 2020 at 8:00 AM Pavithran wrote: > @davisking If I train the model using > tensorflow CNN and I save the weights in dat format....

You aren't running something correct then. I don't know what you are doing wrong though. An entirely self contained script that runs the LFW benchmark correctly is in http://dlib.net/files/dlib_face_recognition_resnet_model_v1_lfw_test_scripts.tar.bz2 On...

The data isn't available. But you don't need to check for this, the model is definitely biased towards generally good looking american celebrities (e.g. white middle aged adults) On Tue,...

You need to make a separate detector for whatever it is you want to find.

I'm not sure if you are trying to detect the positions of things in images, or just classify the whole image as "an A type thing" vs "a B type...

I would try this stuff for finding the positions of things: http://dlib.net/ml.html#loss_mmod_