So I am currently writing a yolo model in Keras, that performs object detection on videos, uses a videowriter which writes to the output frame by frame(with the bounding boxes).
