Please provide complete information as applicable to your setup.
• Hardware Platform (Jetson )
**• DeepStream Version5.1
• JetPack Version (4.5)
**• TensorRT Version7.x
This is my pipeline：
uridecodebin --> nvstreammux --> queue1 --> nvinfer(pgie) --> queue2 --> nvvideoconvert--> queue3 --> capsfilter --> queue4 --> tracker--> queue5 --> nvinfer(pgie2) --> queue9-->fakesink
The first level network is face detection，The secondary network is a facenet model，The input source is two RTSP webcams.The program can run normally，However, when the target appears, the program will automatically exit immediately，This doesn’t happen when I cancel the secondary network or put the tracking model behind the secondary network.