Performance of gst-dsexample deepstream

Please provide complete information as applicable to your setup.

• Hardware Platform (Jetson / GPU) Jetson
• DeepStream Version Deepstream 6.0
• JetPack Version (valid for Jetson only) Jetpack 4.6 rev3
• TensorRT Version 8.0.1

I am using blur_objects of gst-dsexample for pre-processing and I observe a latency. Is it because in gst-dsexample opencv operations are off-loaded to CPU?
I have seen similar issue: Performane of Deepstrem gst-dsexample plugin

How to implement pre-processing in gst-dsexample on GPU seamlessly?

You may consider to implement the blur algorithm with CUDA to accelerate by Nvidia GPU. About CUDA | NVIDIA Developer

Sorry for the late reply but of late I am concerned of RTSP stream of deepstream app to VLC. By using nvvoverlaysink with sync=0 I face very little latency to make the app work but when using RTSP over a VLC media player, I see a lot of latency while visualizing the output of deepstream app. How can I make such visualization hardware accelerated?

RTSP is network transferring. It can not be accelerated by any other HW but your network adapter.

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.

Thank you for the resolution

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