Giter VIP home page Giter VIP logo

raspberrypi_webrtc's Introduction

RaspberryPi_WebRTC

Using v4l2 dma hardware encoder with WebRTC reduces CPU usage. Both signalr and mqtt are the options for signaling in this project.

Architecture

architecture

Environment

  • RaspberryPi 3B + Raspberry Pi Camera v1.3
  • RaspberryPi OS 64bit
  • clang 12+
  • boringssl replace openssl

Summary

  • Latency is about 0.2 seconds delay.
  • Temperatures up to 60~65°C.
  • Using HW DMA sources minimizes CPU usage and latency down to ~50ms.

latency latency


How to use

Preparation

  1. Follow SETUP_ARM64_ENV to prepare an arm64 env for compilation (Optional)
  2. Follow BUILD_WEBRTC to compile libwebrtc.a
  3. Choose a signaling mechanisum
  4. Install the needed packages on pi
    sudo apt install libboost-program-options-dev libavformat-dev libavcodec-dev libavutil-dev libswscale-dev libpulse-dev libasound2-dev libx11-dev
  5. Copy the nlohmann/json.hpp to /usr/local/include
  6. Run signaling server

Compile and run

Command line
Description Valid values
-DUSE_SIGNALR_SIGNALING Build the project by using SignalR as signaling. ON, OFF
-DUSE_MQTT_SIGNALING Build the project by using MOSQUITTO as signaling. ON, OFF
-DBUILD_TEST Build the test codes recorder, mqtt, v4l2_capture, v4l2_encoder, v4l2_decoder, v4l2_scaler
-DUSE_BUILT_IN_H264 Use the built-in openh264 software encoder ON, OFF

Build on raspberry pi and it'll output a pi_webrtc file in /build.

mkdir build
cd build
cmake .. -DCMAKE_CXX_COMPILER=/usr/bin/clang++ -DUSE_MQTT_SIGNALING=ON
make -j

Run pi_webrtc to start the service.

  • If use signalr as signaling.
    ./pi_webrtc --device=/dev/video0 --fps=30 --width=1280 --height=720 --v4l2_format=mjpeg --signaling_url=http://localhost:6080/SignalingServer --enable_v4l2_dma
  • If use mosquitto mqtt as signaling.
    ./pi_webrtc --device=/dev/video0 --fps=30 --width=1280 --height=720 --v4l2_format=mjpeg --mqtt_host=127.0.0.1 --mqtt_port=1883 --mqtt_username=<username> --mqtt_password=<password>  --enable_v4l2_dma
  • ./pi_webrtc -h to list all available args.
  • If --enable_v4l2_dma is assigned, only H264 stream is provided. The VP8, VP9 are available as well if not be assigned. But frames will be decoded/scaled by software, and the buffer will be copied to the HW encoder though.
  • If the --record_path is assigned, the background recorder will start immediately after running the program. But the performance of Pi 3B is limited, if the resolution is above 640x368@15fps the HW encoder will be unstable, stuck, or even crash.

Run as Linux Service

Set pi_webrtc to run as a daemon.

  • Create /etc/systemd/system/webrtc.service, config sample:
    [Unit]
    Description= the webrtc service need signaling server first
    After=systemd-networkd.service farmer-api.service
    
    [Service]
    Type=simple
    WorkingDirectory=/home/pi/IoT/RaspberryPi_WebRTC/build
    ExecStartPre=/bin/sleep 60
    ExecStart=/home/pi/IoT/RaspberryPi_WebRTC/build/pi_webrtc --fps=30 --width=1280 --height=720 --v4l2_format=h264 --enable_v4l2_dma --mqtt_username=hakunamatata --mqtt_password=wonderful --record_path=/home/pi/video/
    Restart=always
    RestartSec=20
      
    [Install]
    WantedBy=multi-user.target
  • Enable and run the service
    sudo systemctl enable webrtc.service
    sudo systemctl start webrtc.service

Install the coturn (Optional)

If the cellular network is used, the coturn is required because the 5G NAT setting by ISP may block p2p.

  1. Install
    sudo apt update
    sudo apt install coturn
    sudo systemctl stop coturn.service
  2. Edit config sudo nano /etc/turnserver.conf, uncomment or modify below options
    listening-port=3478
    listening-ip=192.168.x.x
    relay-ip=192.168.x.x
    external-ip=174.127.x.x/192.168.x.x
    #verbose
    lt-cred-mech
    user=webrtc:webrtc
    realm=greenhouse
    no-tls
    no-dtls
    syslog
    no-cli
  3. Set the port 3478 forwarding on the router
  4. Start the service, sudo systemctl start coturn.service

Reference

raspberrypi_webrtc's People

Contributors

tzuhuantai avatar

Recommend Projects

  • React photo React

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

  • Vue.js photo Vue.js

    🖖 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. 📊📈🎉

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 ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.