/kinesisvideo-encoder-ros2

ROS packages for facilitating the use of AWS cloud services.

Primary LanguageC++Apache License 2.0Apache-2.0

h264_video_encoder

Overview

This package provides a ROS Node that will encode a stream of images into an H264 video stream.

Keywords: ROS, AWS, Kinesis

License

The source code is released under LGPL 2.1. However, this package uses h264_encoder_core which incorporates several different encoding components which may further restrict the license. By default, x264 is used for software encoding, thereby applying GPL to all of h264_video_encoder.

Author: AWS RoboMaker
Affiliation: Amazon Web Services (AWS)
Maintainer: AWS RoboMaker, ros-contributions@amazon.com

Supported ROS Distributions

  • Dashing

Build status

  • Travis CI:
    • "master" branch
    • "release-latest" branch
  • ROS build farm:

Installation

Binaries

Not available in apt yet On Ubuntu you can install the latest version of this package using the following command

    sudo apt-get update
    sudo apt-get install -y ros-$ROS_DISTRO-h264-video-encoder

Building from Source

To build from source you'll need to create a new workspace, clone and checkout the latest release branch of this repository, install all the dependencies, and compile. If you need the latest development features you can clone from the master branch instead of the latest release branch. While we guarantee the release branches are stable, the master should be considered to have an unstable build due to ongoing development.

  • Create a ROS workspace and a source directory

      mkdir -p ~/ros-workspace/src
    
  • Clone the package into the source directory .

    Note: Replace {MAJOR.VERSION} below with the latest major version number to get the latest release branch.

      cd ~/ros-workspace/src
      git clone https://github.com/aws-robotics/kinesisvideo-encoder-ros2.git -b release-v{MAJOR.VERSION}
    
  • Install dependencies

      cd ~/ros-workspace 
      sudo apt-get update && rosdep update
      rosdep install --from-paths src --ignore-src -r -y
    

Note: If building the master branch instead of a release branch you may need to also checkout and build the master branches of the packages this package depends on.

  • Build the packages

      cd ~/ros-workspace && colcon build
    
  • Configure ROS library Path

      source ~/ros-workspace/install/setup.bash
    
  • Build and run the unit tests

      colcon build 
      colcon test && colcon test-results --all
    

Building on Cloud9 - Cross Compilation

  • In RoboMaker's Cloud9, start with an empty workspace and in the Cloud9 console:

     # build docker image
     cd /opt/robomaker/cross-compilation-dockerfile/
     sudo bin/build_image.bash  # this step will take a while
    
     # create workspace
     mkdir -p ~/environment/robot_ws/src
     cd ~/environment/robot_ws/src
     git clone https://github.com/aws-robotics/kinesisvideo-encoder-common.git
     git clone https://github.com/aws-robotics/kinesisvideo-encoder-ros2.git
    
     # run docker image
     cd ..
     sudo docker run -v $(pwd):/ws -it ros-cross-compile:armhf
    
  • Now you're inside the cross-compilation docker container

     # build the workspace
     cd ws
     apt update
     rosdep install --from-paths src --ignore-src -r -y  # this step will take a while
     colcon build --build-base armhf_build --install-base armhf_install
     colcon bundle --build-base armhf_build --install-base armhf_install --bundle-base armhf_bundle --apt-sources-list /opt/cross/apt-sources.yaml   # this step will take a while
     exit
    
  • Now you're oustide the cross-compilation docker container

     # for more on copying s3 buckets see: https://docs.aws.amazon.com/cli/latest/reference/s3/cp.html
     aws s3 cp armhf_bundle/output.tar.gz s3://<bucket_name_in_your_robomaker_account>/h264_video_encoder.armhf.tar
    

Launch Files

A launch file called h264_video_encoder_launch.py is included in this package. The launch file uses the following arguments:

Arg Name Description
node_name (optional) The name the H264 encoder node should be launched with. If not provided, the node name will default to h264_video_encoder
config (optional) A path to a ros2 parameters yaml file.

An example params file is include in config/sample_configuration.yaml.

Usage

Running the node

To launch the H264 encoder node, you can run the following command:

ros2 launch h264_video_encoder h264_video_encoder_launch.py 

Configuration File and Parameters

An example configuration file called sample_configuration.yaml is provided for running the H264 encoder node on a Raspberry Pi based system. When the parameters are absent in the ROS parameter server, default values are used, thus all parameters are optional. See table below for details.

Parameter Name Description Type
queue_size (optional) The maximum number of incoming and outgoing messages to be queued towards the subscribed and publishing topics. integer
output_width (optional) The desired width (in pixels) of each frame in the encoded video output. integer
output_height (optional) The desired height (in pixels) of each frame in the encoded video output. integer
fps_numerator (optional) The desired frames per second (the numerator portion when expressing FPS as a rational number) for the encoded video output. integer
fps_denominator (optional) The desired frames per second (the denominator portion when expressing FPS as a rational number) for the encoded video output. integer
bitrate (optional) The desired bitrate (in bits per second) of the encoded video output. integer

Node Details

Published Topics

Topic Name Message Type Description
Configurable (default="video/encoded") kinesis_video_msgs/KinesisVideoFrame The node will publish to a topic of a given name. Each message being published contains a chunk of the video stream, usually per video frame.

Subscribed Topics

Topic Name Message Type Description
Configurable (default="/raspicam_node/image") sensor_msgs/Image The node will subscribe to a topic of a given name. The data is expected to be a stream of images from a source (such as a Raspberry Pi camera).

Bugs & Feature Requests

Please contact the team directly if you would like to request a feature.

Please report bugs in Issue Tracker.