search

Seanlinx / mtcnn

595 stars 264 forks source link

Quick question regarding threshold and running on batch of images #49

Open srireddy821 opened 7 years ago

srireddy821 commented 7 years ago

  1. What are the recommended thresholds for nets in your framework? parser.add_argument('--thresh', dest='thresh', help='list of thresh for pnet, rnet, onet', nargs="+",default=[0.6, 0.7, 0.7], type=float)

  2. Does below argument sets the net to run on batch of images?

parser.add_argument('--batch_size', dest='batch_size', help='list of batch size used in prediction', nargs="+", default=[2048, 256, 16], type=int)

Seanlinx commented 7 years ago

@srireddy821

  1. I use [0.6, 0.15, 0.05] while testing on FDDB. The thresholds are related to your training process(eg. epochs, number of images in training set).
  2. The input is a single image, [2048, 256, 16] are for pnet, rnet and onet respectively.
srireddy821 commented 7 years ago

Thanks a lot for the response.

  1. Did you get a chance to train your model on the Widerface dataset?