Deploying trained model to real time object recognition from my laptop webcam

Autonomous Machines Jetson & Embedded Systems Jetson Nano
1

I trained a Hand gesture model in Nvidia digits for american sign language dataset. It is also working fine for testing images, But How can i use my laptop webcam to test images in real time instead of browsing the image every single time for testing.
I have seen Inference tutorials for Nvidia Jetson but no such for Laptop Webcam. KIndly share your work in the below comments if someone has already done it.

Thanks

2

Hi,

You can still use jetson_inference for the deployment.
The sample can be compiled on an x86 environment:
https://github.com/dusty-nv/jetson-inference

For Webcam, suppose you can access it with V4L2 interface as an USB camera.
https://github.com/dusty-nv/jetson-inference/blob/35c0d57766f90faddae372ff1d807e9fc001d26b/docs/detectnet-camera.md#running-the-live-camera-detection-demo

$ ./detectnet-camera --camera=/dev/video0     # using PedNet,  V4L2 camera /dev/video0 (1280x720)

It’s recommended to give it a try first.

Thanks.

3

Hey Thanks for the reply. But i am getting cmake error as below for building the repo link in two days to a demo link

CMake Error at CMakeLists.txt:74 (find_package):
Could not find a configuration file for package “OpenCV” that is compatible
with requested version “3.0.0”.

The following configuration files were considered but not accepted:

/usr/share/OpenCV/OpenCVConfig.cmake, version: 2.4.9.1

could you help me in this.

4

Hi,

The sample requires OpenCV 3.x package.
Could you help to install it?

Thanks.

5

So it will not work for opencv 4.1.1 version?. I need to change to 3.x version to run the two days to a demo tutorial?

6

Hi,

You can give it a try.
Suppose it should work by updating the CMakeLists.txt file here:
https://github.com/dusty-nv/jetson-inference/blob/35c0d57766f90faddae372ff1d807e9fc001d26b/CMakeLists.txt#L74

The OpenCV v3.x is the pre-installed version on the Jetson platform so we fix the version in the CMakeLists.txt.

Thanks.