RTSP output choppy frames

Hello,

I’m testing the rtsp output using the python API of Deepstream 6.0. My application is built on the deepstream-test1-rtsp-out sample app. When I ran the sample app on its own the output livestream was fine. but after I added the nvtracker and nvanalytics elements. It became choppy and pixelated:

I have fiddled around with the sync, qos, buffer size and bitrate parameters, and tried adding queue elements between every couple of plugins like this:

    srcpad.link(sinkpad)

    streammux.link(queue1)

    queue1.link(pgie)

    pgie.link(queue2)

    queue2.link(tracker)

    tracker.link(queue3)

    queue3.link(nvanalytics)

    nvanalytics.link(queue4)

    queue4.link(nvvidconv)

    nvvidconv.link(queue5)

    queue5.link(nvosd)

    nvosd.link(queue6)

    queue6.link(nvvidconv_postosd)

    nvvidconv_postosd.link(queue7)

    queue7.link(caps)

    caps.link(queue8)

    queue8.link(encoder)

    encoder.link(queue9)

    queue9.link(rtppay)

    rtppay.link(queue10)

    queue10.link(sink)

But nothing seems to be working. I checked other issues in this forum but couldn’t find any fix. I would really appreciate your help.

• Hardware Platform: Jetson Nano
• DeepStream Version: 6.0
• JetPack Version: 4.6
• TensorRT 8
• Issue Type: Bug
• Steps to Reprodue: Run the python deepstream-nvdsanalytics sample app with the RTSP output from the python deepstream-test1-rtsp-out sample app

From the image you post, there are packets lost happen with your network. You may try to set smaller target bitrate or set shorter “iframeinterval” with nvv4l2h264enc/nvv4l2h265enc. It may help to improve the quality but to improve your network throughputand packet lost rate is more useful than change anything with deepstream.

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