A project demonstrating use of Python for DeepStream sample apps given as a part of SDK (that are currently in C,C++).

DeepStream Python Apps

This repository contains Python bindings and sample applications for the DeepStream SDK.

SDK version supported: 5.1

Download the latest release package complete with bindings and sample applications from the release section.

Please report any issues or bugs on the Deepstream SDK Forums.

• Python Bindings
• Sample Applications

Python Bindings

DeepStream pipelines can be constructed using Gst Python, the GStreamer framework’s Python bindings. For accessing DeepStream MetaData, Python bindings are provided in the form of a compiled module which is included in the DeepStream SDK. This module is generated using Pybind11.

These bindings support a Python interface to the MetaData structures and functions. Usage of this interface is documented in the HOW-TO Guide and demonstrated in the sample applications.
This release adds bindings for decoded image buffers (NvBufSurface) as well as inference output tensors (NvDsInferTensorMeta).

Sample Applications

Sample applications provided here demonstrate how to work with DeepStream pipelines using Python.
The sample applications require MetaData Bindings to work.

To run the sample applications or write your own, please consult the HOW-TO Guide

We currently provide the following sample applications:

• deepstream-test1 – 4-class object detection pipeline
• deepstream-test2 – 4-class object detection, tracking and attribute classification pipeline
• deepstream-test3 – multi-stream pipeline performing 4-class object detection
• deepstream-test4 – msgbroker for sending analytics results to the cloud
• deepstream-imagedata-multistream – multi-stream pipeline with access to image buffers
• deepstream-ssd-parser – SSD model inference via Triton server with output parsing in Python
• deepstream-test1-usbcam – deepstream-test1 pipeline with USB camera input
• deepstream-test1-rtsp-out – deepstream-test1 pipeline with RTSP output
• deepstream-opticalflow – optical flow and visualization pipeline with flow vectors returned in NumPy array
• deepstream-segmentation – segmentation and visualization pipeline with segmentation mask returned in NumPy array
• deepstream-nvdsanalytics – multistream pipeline with analytics plugin

Of these applications, the following have been updated or added in this release:

• deepstream-test2: added option to enable output of past frame tracking data
• deepstream-test4: callback functions are registered only once to avoid race condition
• deepstream-imagedata-multistream: the probe function now modifies images in-place in addition to saving copies of them
• deepstream-opticalflow: new sample application to demonstrate optical flow functionality
• deepstream-segmentation: new sample application to demonstrate segmentation functionality
• deepstream-nvdsnalaytics: new sample application to demonstrate analytics functionality

Detailed application information is provided in each application’s subdirectory under apps.