What Happens When a Stream is Dropped when using NVStreamMux?

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1

• Hardware Platform: GPU
• DeepStream Version: 6.2

I’m building an application that decodes RTSP streams from several different cameras. I then use NVStreamMux to batch these frames. As I can’t really test it right now, and I have experienced in the past certain cameras will have internal errors where they lose connection for several minutes- I’m wondering how losing a camera stream would internally affect the DeepStream pipeline?

Will NVStreamMux just fill it’s batch with frames from other cameras?
When the camera comes back online will NVStreamMux go back to batching 1 frame from each camera?

3

please refer to nvstreammux introduction doc, The muxer uses a round-robin algorithm to collect frames from the sources. It tries to collect an average of (batch-size/num-source) frames per batch from each source. when one source can’t input frames to muxer, muxer will wait to timeout if the complete batch is not formed. please find batched-push-timeout in the line above.