Skip to content
This repository has been archived by the owner on Jun 6, 2019. It is now read-only.

Releases: intel/node-realsense

v0.10.0 Release

31 Mar 10:26
@DonnaWuDongxia DonnaWuDongxia
v0.10.0
637be24
Compare
Choose a tag to compare
v0.10.0 Release Pre-release
Pre-release

This release includes three algorithm modules and a basic library to access Intel ® RealSense TM Technology for Linux OS in Node.js context. While librealsense module providing interfaces to access the raw data from 3D camera, three algorithm modules, including Person Tracking, Object Recognition and SLAM can be used to do practical applications in specifica areas. Samples for the algorithm modules are also included in this release to demonstrate how to usage these modules in JavaScript style.

Release Note

  • Multiple model support: with this release, it's able to run arbitrary number of modules (node-object, node-person and/or node-slam) at the same time.
  • Object Library: no significant change for the module itself, but it have been linked with other algorithm modules by multi-model support.
  • Person Library: Person Recognition was supported in Person Tracking module, that can be used to register and recognize persons.
  • SLAM: Relocation Map features, reset and restart control interfaces are all ready in this release.
  • Librealsense module: get basic information of the device, set the parameters of the video stream and get the data.
  • Known issues:
    • GPU work mode of Object Library might not work on Joule.
    • Memory leak in Object Library has been found.
    • Librealsense Library has not been linked in the multi-module usage.
    • No Librealsense Library based samples currently.
    • Not all designed interfaces for Librealsense Library have been implemented yet.
    • There is a bug(#194) on multi-module feature, application aiming to combine SLAM with other two modules needs to set customized camera options as this example. What's more, please avoid to start and stop the camera multiple times in a single process. We are working to fix it.

Environment Setup

Getting started doc is a good start point.
Setup Intel® RealSense™ with Ubuntu 16.04 on Intel® Joule™.

Install Node.js and NPM for Ubuntu

sudo apt-get update
sudo apt-get install nodejs npm

# Workaround of node not found
sudo ln -s /usr/bin/nodejs /usr/bin/node

# Workaround of boost library missing
sudo apt install libboost-thread-dev

# Install modules in the dir you're working, choose the package you want
npm i --save node-object@0.10.0 node-person@0.10.0 node-slam@0.10.0 node-librealsense@0.10.0

Connect the Intel® RealSense™ ZR300 camera to the USB 3.0 port of Joule.

Examples

This Github Repo contains Person Tracking, Object Recognition and SLAM examples using the above Node.js modules/addons.

# Object Library Tutorial No.1
cd samples/or_tutorial_1/javascript
npm i
node main.js

# Person Library Tutorial No.1
cd samples/pt_tutorial_1/javascript
npm i
node main.js

# SLAM Library Tutorial No.1
cd samples/slam_tutorial_1/javascript
npm i
node main.js

API Specs

  • Person Tracking JavaScript API spec is located at /path/to/node-realsense-0.10.0/doc/spec/person-tracking.html.
  • Object Recognition JavaScript API spec is located at /path/to/node-realsense-0.10.0/doc/spec/object-recognition.html
  • SLAM JavaScript API spec is located at /path/to/node-realsense-0.10.0/doc/spec/slam.html
  • The spec of common API in all three modules is located at /path/to/node-realsense-0.10.0/doc/spec/common.html
  • The designed API of Librealsense module is located at /path/to/node-realsense-0.10.0/dsrc/librealsense/apis.md
Assets 7

v0.9.1 Release

27 Feb 07:20
@kenny-y kenny-y
v0.9.1
cb06c12
Compare
Choose a tag to compare
v0.9.1 Release Pre-release
Pre-release

This release includes the Person Tracking, Object Recognition and SLAM JavaScript API of the Intel ® RealSense TM Technology for Linux OS. JavaScript samples for Person Tracking, Object Recognition and SLAM.

Release Note

  • Object Library: Object Recognition, Object Localization, and Object Tracking features are ready to be used.
  • Person Library: Person Detection, Person Tracking, Skeleton Recognition and Gesture Recognition features are ready to be used.
  • SLAM: Camera Pose Tracking and Occupancy Mapping features are ready to be used.
  • Known issues:
    • GPU work mode of Object Library might not work on Joule.
    • Memory leak in Object Library has been found.
    • Person Recognition feature in Person Library is not supported in this version.
    • Re-localization in SLAM is not supported in this version.

Environment Setup

Setup Intel® RealSense™ with Ubuntu 16.04 on Intel® Joule™.

Install Node.js and NPM for Ubuntu

sudo apt-get update
sudo apt-get install nodejs npm

# Workaround of node not found
sudo ln -s /usr/bin/nodejs /usr/bin/node

# Workaround of boost library missing
sudo apt install libboost-thread-dev

# Install modules in the dir you're working, choose the package you want
npm i --save node-object@0.9.2 node-person@0.9.1 node-slam@0.9.1

Connect the Intel® RealSense™ ZR300 camera to the USB 3.0 port of Joule.

Examples

This Github Repo contains Person Tracking, Object Recognition and SLAM examples using the above Node.js modules/addons.

# Object Library Tutorial No.1
cd samples/or_tutorial_1/javascript
npm i
node main.js

# Person Library Tutorial No.1
cd samples/pt_tutorial_1/javascript
npm i
node main.js

# SLAM Library Tutorial No.1
cd samples/slam_tutorial_1/javascript
npm i
node main.js

API Specs

  • Person Tracking JavaScript API spec is located at /path/to/node-realsense-0.9.1/doc/spec/person-tracking.html.
  • Object Recognition JavaScript API spec is located at /path/to/node-realsense-0.9.1/doc/spec/object-recognition.html
  • SLAM JavaScript API spec is located at /path/to/node-realsense-0.9.1/doc/spec/slam.html
  • The spec of common API in all three modules is located at /path/to/node-realsense-0.9.1/doc/spec/common.html
Assets 6