Hello everybody.
I’m a beginner in DeepStream SDK.
First Question.
I was wondering about that if i set a Gst-nvinfer plugin in primary mode,then set a Gst-nvtracker plugin behind it,(assumed that all the identified objects are tracked effectively)will the Gst-nvinfer plugin tries to improve performance by avoiding re-inferencing on the same objects in every frame?
I found NVIDIA docs and said it was supported when the nvinfer work as Secondary mode.
Does the primary mode support it?
Second Question.
By the way, if i got a rtmp stream in 30 FPS, Could i set nvinfer to do inference one frame per sec and rest of the 29 frames use nvtracker to track id? I hope that can improve system performance.