Git Product home page Git Product logo

levipereira / deepstream-yolo-triton-server-rtsp-out Goto Github PK

View Code? Open in Web Editor NEW
4.0 4.0 1.0 40 KB

The Purpose of this repository is to create a DeepStream/Triton-Server sample application that utilizes yolov7, yolov7-qat, yolov9 models to perform inference on video files or RTSP streams.

License: Apache License 2.0

Python 100.00%
deepstream deepstream-python deepstream-python-apps deepstreamsdk triton-inference-server triton-server yolov7 yolov9

deepstream-yolo-triton-server-rtsp-out's Introduction

Sample App Nvidia DeepStream - Triton-Server (YOLO Models)

The Purpose of this repository is to create a DeepStream/Triton-Server sample application that utilizes yolov7, yolov7-qat, yolov9 yolov9-qat models to perform inference on video files or RTSP streams. It then showcases the output on an RTSP URL, providing a straightforward demonstration of end-to-end AI processing.

Follow this steps to Use this Sample App

1. Deploy and Start Triton Server

Follow the steps on below link to Start Triton Server
Triton Server - YOLO.

2. Deploy and Start DeepStream

  1. Start Nvidia Container nvcr.io/nvidia/deepstream:6.4-triton-multiarch
    Install Deepstream Python Bindings ./user_deepstream_python_apps_install.sh --version 1.1.10

  2. Install custom parse lib NvDsInferYolov9EfficientNMS for Gst-nvinferserver
    Clone this repository nvdsinfer_yolo_efficient_nms
    The custom library is built and installed using the provided Makefile.

  3. Install Application for DeepStream 6.4 ds-6.4-ts-yolo-rtsp-out Follow the steps on below link to Build DeepStream
    DeepStream - YOLO

  4. Using Sample Application ds-6.4-ts-yolo-rtsp-out

    To use this sample application, follow the steps below from within the container:

    The sample application is installed on /deepstream_python_apps/apps/ds-6.4-ts-yolo-rtsp-out/

    This sample application supports input streaming from both file:// and rtsp:// sources. It processes the input and streams the output via RTSP at the following URL: rtsp://localhost:8554/ds-test

    python3 ds-6.4-ts-yolo-rtsp-out.py -i <input_files> -m <model> -c <codec> -b <bitrate> [--rtsp-ts]
    
    #example
    python3 ds-6.4-ts-yolo-rtsp-out.py  \
    -i file:///opt/nvidia/deepstream/deepstream/samples/streams/sample_1080p_h264.mp4  \
    -m  yolov9-c \
    -c H264   

    Arguments

    • -i, --input: Path to input file:// or rtsp:// elementary stream. Multiple input files can be provided.
    • -m, --model: Choice ['yolov7','yolov7x','yolov7-qat','yolov7x-qat','yolov9-c','yolov9-e', 'yolov9-c-qat','yolov9-e-qat'].
    • -c, --codec: RTSP Streaming Codec. Choose between H264 and H265. (Default: H264)
    • -b, --bitrate: Set the encoding bitrate. (Default: 4000000)
    • --rtsp-ts: Attach NTP timestamp from RTSP source. (Default: False)

deepstream-yolo-triton-server-rtsp-out's People

Contributors

levipereira avatar

Stargazers

 avatar  avatar  avatar  avatar

Watchers

 avatar

Forkers

soon14

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.