leo7r/swiftface

How to use it? any demo code for python?

Opened this issue · 3 comments

Hi,
how to implement your code, i need to test that using a web cam and find out the accuracy

leo7r commented

Hi, the code is just a python implementation of the darknet code (if you need a tutorial, this is a good starting point https://towardsdatascience.com/yolo-object-detection-with-opencv-and-python-21e50ac599e9). What you need to do after getting it working is to use swiftface.weights when doing inference.

I'm working on a Google Colab so people can test more quickly, let you know when ready.

Great thanks for replaying to me, i have already implemented something that you have shared in the link ,,,,also the link is very helpful.... now if possible that share some code for this part "use swiftface.weights when doing inference." it will be helpful...i am doing thesis and need to test your code to refer that as well in my study as well

leo7r commented

I found a tutorial for testing darknet on Google Colab https://towardsdatascience.com/object-detection-using-yolov3-on-colab-5d7d9eef02b3.

The only difference is that instead of downloading weights in Step #5
6A55617D-09B3-4C1A-8ADD-E85F1E2560BE you should use the weights on SwiftFace repo.