Multiple camera streams with the jetson-inference example

[lars2]

Hi,

I’ve been successfully testing the jetson-inference DetectNet c++ application (./detectnet) with a rtp streaming camera.
Also managed to write my own version based on the my-recognition example. How can I add support for more than one camera? And would this require multiple instances of the detection network or can I feed the images through the same instance?

Also what is the difference between jetson-inference and using DeepStream SDK or other alternatives?

Thanks for all help,
Lars

[DaneLLL]

Hi,
For multiple RTSP sources, it might be easier to use DeepStream SDK. You can modify config file and run deepstream-app. Please check the latest release:

Announcing DeepStream 5.0.1

Announcing availability of DeepStream 5.0.1 on Devzone and NGC. DeepStream 5.0.1 fixed few bugs that our users found in DeepStream 5.0. Here’s a list of the bugs that we addressed in DeepStream 5.0.1 Demuxer plugin – Stream content were mixing after nvstreamdemux Dewarper plugin – Fixed aspect ratio Python binding – Fixed bindings for NvDsLabelInfo The Demuxer plugin bug has been reported by several users; please upgrade to DeepStream 5.0.1 to pick up the latest fixes. Here are few issues re…

[dusty_nv]

@lars2

Also managed to write my own version based on the my-recognition example. How can I add support for more than one camera?

You can create additional jetson-inference videoSource objects that use different cameras. However, the command-line parsing is setup for one camera. So you may wish to use this videoSource::Create() constructor where you manually pass in the video options and camera URIs:

https://rawgit.com/dusty-nv/jetson-inference/dev/docs/html/classvideoSource.html#ac84b9d20c628c19e0463aca50ed2cdf9

And would this require multiple instances of the detection network or can I feed the images through the same instance?

You can feed multiple sources through one instance of imageNet/detectNet/ect.