Skip to content

Latest commit

 

History

History
76 lines (48 loc) · 2.79 KB

README.md

File metadata and controls

76 lines (48 loc) · 2.79 KB
Raw

Deepstream People Trajectory CCTV

This project utilizes NVIDIA's DeepStream SDK to track the trajectory of people in a CCTV feed. The application processes video streams, detects objects, and visualizes their paths in real-time. It can be particularly useful for analyzing foot traffic in retail environments or monitoring public spaces.

Sample Output

Overview

The application draws the trajectory paths of people walking through a defined region of interest (ROI). It can help identify areas with high demand, such as locations near advertisement hoardings or popular sections in a store.

Use Cases

  • Retail Analysis: Understand customer movement patterns and optimize store layouts.
  • Security Monitoring: Track individuals in sensitive areas for enhanced security.
  • Traffic Management: Analyze pedestrian traffic in public spaces.

Requirements

  • Python 3.x
  • GStreamer
  • NVIDIA DeepStream SDK
  • Required Python packages (e.g., gi, pyds, argparse, configparser)

Installation

  1. Clone the repository:

    git clone <repository-url>
cd Deepstream_people_trajectory_CCTV

  2. Install the required dependencies:

    pip install -r requirements.txt

  3. Ensure you have the NVIDIA DeepStream SDK installed and configured on your system.

Configuration

The application requires configuration files for the primary and secondary inference engines, as well as the tracker. These files should be placed in the project directory:

  • dstest2_pgie_config.txt: Configuration for the primary inference engine.
  • dstest2_tracker_config.txt: Configuration for the tracker.

Model Information

The models used in this project can be found in the model folder. Ensure that the model files are compatible with the DeepStream SDK and are properly configured in the respective configuration files.

Usage

To run the application, use the following command:

python trajectory.py -i <input_stream>

Replace <input_stream> with the path to your input H264 elementary stream or RTSP URL. You can specify multiple input streams by separating them with spaces.

Command Line Arguments

  • -i, --input: Path to input H264 elementary stream (required).
  • -g, --gie: Choose GPU inference engine type (nvinfer or nvinferserver, default is nvinfer).
  • -c, --codec: RTSP Streaming Codec (H264 or H265, default is H264).
  • -b, --bitrate: Set the encoding bitrate (default is 4000000).
  • --rtsp-ts: Attach NTP timestamp from RTSP source (optional).

License

This project is licensed under the Apache License, Version 2.0. See the LICENSE file for details.

Acknowledgments

  • NVIDIA for the DeepStream SDK.
  • The open-source community for various libraries and tools used in this project.