Why there is delay streaming with rtsp from a dvr in deepstream?

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1

I am running peoplenet using Xavier and streaming from a dvr using rtsp.
I am streaming 7 cctv.
The longer the device runs, delay is getting longer.
I use deepstream-app for my application.

If I stream fewer CCTVs, delay is shorter. But there is delay.
How can I remove the delay totally? I can drop some frames, I like to get latest images when streaming.

My device infos are

• Hardware Platform (Jetson / GPU) Xavier
• DeepStream Version 6.0 (latest)
• NVIDIA GPU Driver Version (valid for GPU only)
10.2

Config files are attached.
config_infer_stluke.txt (4.1 KB)
stluke_1080p_dec_infer-resnet_tracker_sgie_tiled_display_int8_record.txt (8.2 KB)

3

live-stream is set 1 and sync=0 is set. Then Xavier can stream 4 cctv in realtime.
When increased to 6 cctvs, there is 5 sec delay compared to DVR.
4 cctvs are enough for my application.
thanks

4

Is this issue resolved?

5

In this test, AGX Xavier’s performance is 305 FPS for Int8 peoplenet model.
I am using the same model and only additional process is tracking.
Why I have 15 FPS with 4 CCTVs, so my application achieved 60FPS only.
Why so much difference from performance test?

6

Did you boost the system by nvpmodel and jetson_clocks?

sudo nvpmodel -m 0
sudo jeston_clocks

7

I’m closing this topic due to there is no update from you for a period, assuming this issue was resolved.
If still need the support, please open a new topic. Thanks

8

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