search

GilLevi / AgeGenderDeepLearning

633 stars 285 forks source link

Human Vs Animal problem #36

Open macd2 opened 5 years ago

macd2 commented 5 years ago

Hey There, first of all thanks for this great examples. While testing i found some really bad false positive it looks like the net can not distinguish between a human and an animal.

Here are 2 examples:

  1. https://i.imgur.com/6FaLr7w.jpg
  2. https://i.imgur.com/ImVjWsc.jpg

What is the best way to go about this? Or you plan to release any updated models in the near future? thanks

GilLevi commented 5 years ago

Hi,

Thank you for your interest in our work.

The common pipeline for using the network is applying face detection on the image and feeding the detected network (ideally, after alignment) to the network.

Best, Gil

macd2 commented 5 years ago

Hey Gil, thanks for the replay. I'm not sure if i understand...

So what do you suggest doing in the above described case?

GilLevi commented 5 years ago

In the above case, I suggest running a (human) face detector and run the network on the faces it detected in the image. Since it won't probably detect the dog faces, you won't run the network on that.

macd2 commented 5 years ago

@GilLevi ok thanks for that! are you planning to release any updates on this network at all?

GilLevi commented 5 years ago

Not in the near future.