TensorRT Detectron2 webcam

AI & Data Science Deep Learning (Training & Inference) TensorRT
1

Description

Hello Everyone!
So after a lot of tries, I was finally able to develop a TensorRT engine for the detectron2’s mask-rcnn in the docker.
While I did do inference to check the performance, I was wondering if you guys could advice me something.

I want to do real-time inference with a webcam so if you know anything, it would be a big help.
Of course, I am trying to develop the code myself too but it would be a big help if you could perhaps point me to the right direction.

I am also simultaneously doing it on Jetson now that I have updated my jetpack to 5.1.

Thank you all!

Environment

TensorRT Version: 8.5.3.1
CUDA Version: 11.8
CUDNN Version: 8.6
Operating System + Version: Ubuntu20.04
Python Version (if applicable): 3.8
PyTorch Version (if applicable): 2

Relevant Files

TensorRT for Detection

Docker Image for Jetpack 5.1

2

Hi,

This looks like a Jetson issue. Please refer to the below samples in case useful.

For any further assistance, we will move this post to to Jetson related forum.

Thanks!

4

I apologize if I did not make myself clear,
I already created the tensorrt engine for detectron2.

I just want to run inference using a webcam so I was asking for ideas here.

5

Hi @rajupadhyay59 ,
Please refer to the link for understanding and performing inference on TRT engine.

Thanks

6

@AakankshaS Thank you for the reply, but I already solved the problem.
Thanks, everyone!

1 Like