Hello Object Detection

Get jetson-inference repository

$ cd
$ git clone --recursive https://github.com/dusty-nv/jetson-inference

Create workspace

$ mkdir jetson_ws

Run the container

Go inside the jetson-inference repository

$ cd ~/jetson-inference
$ docker/run.sh --volume /home/jetson/jetson_ws:/jetson-inference/jetson_ws

Code

helloai.py

import jetson.inference
import jetson.utils
 
net = jetson.inference.detectNet("ssd-mobilenet-v2", threshold=0.5)
camera = jetson.utils.gstCamera(1280, 720, "/dev/video0")
display = jetson.utils.glDisplay()
 
while display.IsOpen():
	img, width, height = camera.CaptureRGBA()
	detections = net.Detect(img, width, height)
	display.RenderOnce(img, width, height)
	display.SetTitle("Title")

Re-build the container or build your own

Set NVIDIA Container Runtime for Docker

in the file /etc/docker/daemon.json set the default-runtime to nvidia by adding the following json code:

{
    "runtimes": {
        "nvidia": {
            "path": "nvidia-container-runtime",
            "runtimeArgs": []
        }
    },
 
    "default-runtime": "nvidia"
}

Build the container

$ docker/build.sh