How to apply custom object detector on pre-captured video?

Question

Based on what is written on the TensorFlow web page about object-detection API, I have trained a custom model, and its checkpoint is obtained successfully, now I am going to apply this trained model on a pre-captured video(not webcam frames). Now, I would like to know how can I do this? is there any pre-written code that can help me in this way? thanks for your comments. best wishes to you.

Answer 1

You can use most of the code from the docs here.

The only minor change you need to make here is, instead of cap = cv2.VideoCapture(0), use cap = cv2.VideoCapture(path_to_video_file) and it should work