Way to adapt rtsp in- rtsp out deepstream example to output webrtc instead

• Hardware Platform (Jetson / GPU) Jetson Orin Nano
• DeepStream Version 7.0
• JetPack Version (valid for Jetson only) 6.0
• TensorRT Version 8.6.2

I made a program following this example deepstream_python_apps/apps/deepstream-rtsp-in-rtsp-out/deepstream_test1_rtsp_in_rtsp_out.py at master · NVIDIA-AI-IOT/deepstream_python_apps · GitHub in order to read rtsp streams and output rtsp stream so my client can view it. There is a need to make a web app in which they want to include the streams but we are having trouble finding good web components that render rtsp streams directly. So we have 2 choices:

  • Convert rtsp stream to webrtc format (or other suitable format) using another program
  • Modify the rtsp-in rtsp-out example to output the format we need directly avoiding extra programs

I wonder which solution is easier and leads to less complexity and more perfomance.

Any recommendations, suggestions or materials that might help we solve our problem?

Obs: I altered bounding box colors and put some displays texts using nvosd and pyds (not sure if it matters but I thought I should point it out)

There is no webRTC implementation in DeepStream SDK. Even webrtcbin is supported by GStreamer, it is not easy to implement webRTC functions with GStreamer. There are some open source projects which implement webRTC functions with GStreamer.
E.G. imdark/gstreamer-webrtc-demo
Adding WebRTC to your GStreamer application | RidgeRun

1 Like

Thanks for the reply! I was able to make a flask app that displays the rtsp stream after enconding the frames as jpeg with OpenCV. Not sure yet if it is easy to integrate it with my client tech stack, but it is good to know that converting the rtsp format to WebRTC is not the best way.

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.