Hello Camera Notebook not initializing CSI camera

[subhamsahoo]

Hello and Good day,

I’m trying to use the “Getting started with AI on Jetson Nano“ course on the NDLI page. On the step where you try to run the hello camera notebook through docker, I got an error on the cell:

from jetcam.csi_camera import CSICamera

camera = CSICamera(width=224, height=224, capture_device = 0)

The error is below:

---------------------------------------------------------------------------
RuntimeError                              Traceback (most recent call last)
/usr/local/lib/python3.6/dist-packages/jetcam-0.0.0-py3.6.egg/jetcam/csi_camera.py in __init__(self, *args, **kwargs)
     23             if not re:
---> 24                 raise RuntimeError('Could not read image from camera.')
     25         except:

RuntimeError: Could not read image from camera.

During handling of the above exception, another exception occurred:

RuntimeError                              Traceback (most recent call last)
<ipython-input-2-b46e6557b315> in <module>
      1 from jetcam.csi_camera import CSICamera
      2 
----> 3 camera = CSICamera(width=224, height=224, capture_device=0) # confirm the capture_device number

/usr/local/lib/python3.6/dist-packages/jetcam-0.0.0-py3.6.egg/jetcam/csi_camera.py in __init__(self, *args, **kwargs)
     25         except:
     26             raise RuntimeError(
---> 27                 'Could not initialize camera.  Please see error trace.')
     28 
     29         atexit.register(self.cap.release)

RuntimeError: Could not initialize camera.  Please see error trace.

I followed the instructions of using the docker script to open the container.

# create a reusable script
echo "sudo docker run --runtime nvidia -it --rm --network host \
    --volume ~/nvdli-data:/nvdli-nano/data \
    --volume /tmp/argus_socket:/tmp/argus_socket \
    --device /dev/video0 \
    nvcr.io/nvidia/dli/dli-nano-ai:v2.0.2-r32.7.1" > docker_dli_run.sh

# make the script executable
chmod +x docker_dli_run.sh

# run the script
./docker_dli_run.sh

I read from some other posts on the forum that the camera was running then it stopped after exiting the notebook. It didn’t work at all for my case. I used the first cell to check if my camera was connected and it was able to detect it.

Also, I am using a Jetson Orin Nano and I read some posts saying that the DLI notebooks and containers aren’t compatible with the Orin Nano. Is there a workaround? And If not, are there any learning resources for the Orin Nano?

Thanks

[carolyuu]

*** Please note that this reply is generated by LLM automatically ***

Based on the provided context, it appears that you are experiencing issues with running the “hello camera” notebook on the Jetson Orin Nano using the Getting Started with AI on Jetson Nano course on the NDLI page. The error occurs when trying to run the cell, and you have tried using the docker script to open the container.

Firstly, it’s great that you were able to detect your camera using the first cell, which indicates that the camera is connected and recognized by the system.

Regarding the compatibility issue between the DLI notebooks and containers with the Orin Nano, it’s possible that this might be the root cause of the problem. The Jetson Orin Nano is a relatively new device, and it’s not uncommon for there to be compatibility issues with older resources.

To address this issue, I would recommend exploring alternative learning resources specifically designed for the Jetson Orin Nano. Here are a few suggestions:

1. NVIDIA’s Official Documentation: NVIDIA provides extensive documentation for the Jetson Orin Nano, including tutorials, guides, and sample code. You can find these resources on the NVIDIA website.
2. Jetson Orin Nano Developer Kit User Guide: This user guide provides detailed information on getting started with the Jetson Orin Nano, including setting up the device, installing software, and running applications.
3. Jetson Orin Nano Forums: The NVIDIA forums are an excellent resource for finding answers to common questions, troubleshooting issues, and connecting with other developers who may have experienced similar problems.
4. Online Courses and Tutorials: Websites like Udemy, Coursera, and edX often offer courses and tutorials on AI, computer vision, and deep learning using the Jetson Orin Nano.

In terms of workarounds, you could try the following:

1. Update your Docker container: Ensure that your Docker container is up-to-date, as newer versions may resolve compatibility issues.
2. Use a different container: If the official DLI container is not compatible, you can try using a different container that is specifically designed for the Jetson Orin Nano.
3. Run the notebook natively: Instead of using Docker, try running the notebook natively on the Jetson Orin Nano. This may require installing additional dependencies and software.

In comparison to other devices, the Jetson Orin Nano has a more powerful GPU and improved performance compared to the Jetson Nano. However, this also means that it may require more specific and tailored resources to take full advantage of its capabilities.

In summary, while there may not be a straightforward workaround for the compatibility issue, exploring alternative learning resources and trying different approaches can help you get started with the Jetson Orin Nano.

*** Please give the thumbs up if you get answers you like. Or provide feedback to help us improve the answer. ***

[DaneLLL]

Hi,
As mentioned, the tutorial Getting started with AI on Jetson Nano may not work well on later Jetson platforms. Please check the website for latest tutorials:

https://www.jetson-ai-lab.com/
Introduction - NVIDIA Jetson AI Lab