Code example for video inference.

Hello everyone,

I have tensorrt optimized yolov5 engine but I could not run the inference on a video. I checked your ```doInference()``` function but does not match with the one that I used for inference. Since I have little knowledge about cuda programming I did not understand the code. Have you got any reference for using yolo engine on a video.

BEST REGARDS 