Frames as an input for Deepstream python app

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1

Jetson nano Information:
L4T 32.5.1 [ JetPack 4.5.1 ]
Ubuntu 18.04.5 LTS
Kernel Version: 4.9.201-tegra
CUDA 10.2.89
CUDA Architecture: 5.3
OpenCV version: 4.1.1
OpenCV Cuda: NO
CUDNN: 8.0.0.180
TensorRT: 7.1.3.0
Vision Works: 1.6.0.501
VPI: ii libnvvpi1 1.0.15 arm64 NVIDIA Vision Programming Interface library
Vulcan: 1.2.70

Sir, Till now we used file or rtsp stream as an input to the deepstream python test app. Now we want to use frames as an input to the deepstream app instead of video.
Is there any possible way or any such plugin to take frames as an input to the deepstream app.

Thanks in advance.

3

That depends on which kind of frames.

There are lots of image plugins in public Gstreamer. Plugins

4

As we are using video as an input for deepstream python test app, now we want to use np array image frame as an input buffer. Is there any such plugin to achieve that issue. If yes, please kindly share the source code.
Thanks in advance.

5

You can google by yourself. E.G. python - Receive Numpy Array Realtime from GStreamer - Stack Overflow

Or you can refer to How to get numpy array frame from deepstream 5 - Intelligent Video Analytics / DeepStream SDK - NVIDIA Developer Forums

7

Thanks for your quick response. But we do not want to extract numpy array from deepstream app. Our question was to use numpy array which we are getting from third party server as an input to deepstream app instead of video file or rtsp url. Is there any way to do that.

8

Please google by yourself. How to use Gstreamer AppSrc in Python - LifeStyleTransfer

9

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