Are there preconfigured libraries for video object recognition, in which I could pass in an mp4 into a function

Question

I have a video of me walking down a city street, can I pass that into a function that will then print what it sees (telephone pole, trash can, dog, ..ect)? What library could I use?

The key is I dont want to have to train the model (ie pass in a bunch of photos of trach cans or dogs ect)

(ideally python)

SOLVED This article works great actually

https://towardsdatascience.com/how-to-create-a-simple-object-detection-system-with-python-and-imageai-ee1bcaf6b111

Answer 1

yes there are possibilities to do object detection in real time. For instance, YOLO in combination with darknet would maybe be a possibility to for you. For more information see: https://pjreddie.com/darknet/yolo/ and https://github.com/pjreddie/darknet/wiki/YOLO:-Real-Time-Object-Detection