Jetson Nano Inferencing to Mobile App

Hello, I am currently working on a project that requires the jetson nano to send its inferences over to a mobile app(react-native). I want to take a live image classification algorithm(not object detection) where an entire frame is categorized in a class from a webcam feed(v4l2 protocol). I want the results from the nano to be periodically updated every 5 seconds on the mobile app, and I would like the video feed to show as well. Please note that the user needs to see the results on the mobile phone as soon as possible(a low-latency option would be best). I would like to know the best tools to use for this implementation.
I would really appreciate any help!
Thanks

Hi,

Please check our Deepstream SDK first:
https://docs.nvidia.com/metropolis/deepstream/dev-guide/text/DS_Overview.html#deepstream-graph-architecture

It can support RTSP input as well as Kafka, MQTT and AMQP message output.

Thanks.

Thanks so much! I will check it out!

This topic was automatically closed 2 days after the last reply. New replies are no longer allowed.