My computer
Operating system：ubuntu20.04
CPU：i9 Intel(R) Core™ i9-12900KF
GPU： Tesla A10 24G video memory
Deepstream6.1 was installed successfully.
Python successfully bound DeepStream.
Apply Tensorrt yoloV4 FP16
Refer to Gstreamer/RtspSrc_RtspSink_GstBin_pgei.py at main · ahsan-raazaa/Gstreamer · GitHub
made some simple changes to the program。
muti_pushtest.py (29.3 KB)
Input 10-channel RTSP camera，but The RTSP stream received with VLC has a large delay？
How did you measure the delay? What is the CPU loading and GPU loading while you run your python app?
There is time on each frame of the camera, and it does not correspond to the actual time. The difference is 5 minutes, so real-time detection cannot be achieved.
python3 muti_pushtest.py rtsp://admin:Kinghoo123@192.168.1.18/h264/ch35/main/av_stream rtsp://admin:Kinghoo123@192.168.1.18/h264/ch36/main/av_stream rtsp://admin:Kinghoo123@192.168.1.18/h264/ch37/main/av_stream rtsp://admin:Kinghoo123@192.168.1.18/h264/ch38/main/av_stream rtsp://admin:CDWAPM@192.168.1.19/h264/ch1/main/av_stream rtsp://admin:FPTKSH@192.168.1.8/h264/ch1/main/av_stream rtsp://admin:Kinghoo123@192.168.1.18/h264/ch33/main/av_stream rtsp://admin:Kinghoo123@192.168.1.18/h264/ch34/main/av_stream rtsp://admin:TYPCUV@192.168.1.21/h264/ch1/main/av_stream rtsp://admin:LYRQYL@192.168.1.9/h264/ch1/main/av_stream
10 rtsp start run.
The temperature of the graphics card is too high
There is no update from you for a period, assuming this is not an issue anymore.
Hence we are closing this topic. If need further support, please open a new one.
Thanks
Yes. The GPU loading is too high to make the processing realtime. You need to reduce the work load.
