How can i do real time video prediction using TensorRT model ?

I am using Jetson nano for video prediction using tensorflow and opencv, previously i converted my keras model to TensorRT model to be used in Jetson nano.Now i have the TensorRT model so how can i perform real time predicition?

hello thehidden325,

could you please also share what’s your expectation performance of real-time prediction.
you might also refer to Topic 1050377 for Jetson-Nano deep learning inference benchmarks.