pi-webrtc
Configuration, camera backends, signaling, recording and build instructions for the Raspberry Pi and Jetson daemon.
Guides
Quick Start
Everything needed to get a stream running, from a blank SD card to a picture in the browser. Check out the tutorial video for a walkthrough.
Configuration
Every option can be given either as a command-line flag or as a key in a YAML config file. Command-line arguments always win over the file.
Camera and Encoding
--camera takes a <backend>:<id> string. Which backends exist depends on the platform the binary was built for, which CMake detects from /etc/nv_tegra_release...
Signaling
Before two WebRTC peers can send media they have to exchange an SDP offer/answer and a set of ICE candidates. pi-webrtc can do that over three signaling tran...
Recording
Recording is controlled by two independent options. --record-type decides what is captured and --record-mode decides when:
Advanced Usage
An SFU lets one device stream to many viewers without sending a separate stream to each viewer. The SFU receives the stream from the device and forwards it t...
Architecture
pi-webrtc is one process that pulls frames from a camera once and fans them out to three consumers: the WebRTC encoder, the recorder, and — in the sponsor bu...
Server Setup
Build from Source
Build the project
1. Download the prebuilt libwebrtc.a from libwebrtc-builder and install it to /usr/local. To compile it yourself instead, follow BUILD_WEBRTC.
Build the native WebRTC lib (libwebrtc.a)
sudo apt install python3 build-essential libncurses5 libx11-dev
Build the mosquitto lib (mosquitto.a,mosquitto.so and header files )
sudo apt-get install build-essential curl git cmake ninja-build golang libpcre3-dev zlib1g-dev libcjson-dev
Using Google Performance Tools for Performance Testing
This document provides a guide on how to use Google Performance Tools to test performance on a Raspberry Pi and generate performance analysis reports.