Model to train face recognition on NGC

Accelerated Computing Intelligent Video Analytics TAO Toolkit
1

Hi,

I am trying to create an app using DS for face recognition and using Nvidia TLT to train the model.
I am using pretrained VGG16 that is available on NGC.

  1. Can you point me to a better model available on NGC dedicated for face recognition .
  2. Can we train pretrained models available outside of NGC with Nv TLT
2

Hi shkoli,

  1. Firstly,the pre-trained models at ngc are not dedicated for face recognition. Suggest using detectnet_v2 to train a tlt model, which can get a better combination of mAP and FPS.
  2. Yes, actually after you run tlt-train, you already get a tlt format pre-trained model.