Skip to content
openFrameworks addon for the Intel RealSense SR300 depth camera
C++ Makefile
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
example-simple
src
.gitignore
README.md
addon_config.mk
readme_addDirs.png

README.md

#ofxRealSense

Active development, YMMV, things are bound to change...

##Description

ofxRealSense is an openFrameworks addon for the Intel RealSense SR300 depth camera.
It supports 1080p RGB and 480p depth streams, and 3D face scanning. Soon, it will support 3D face tracking, and hopefully skeleton tracking.
Tested on Windows 10, openFrameworks 0.9.8, Visual Studio Community 2015.

Windows only!
Requires Intel RealSense SDK 2016 R2 (v10.0)

###Installation:

Download and install Intel RealSense SDK 2016 R2 here
direct download of 1.8GB exe installation file

Note: Intel RealSense SDK 2016 R3 (v11.0) was released Dec. 2016, which removed face tracking, person tracking, and other modules from the SDK.
As such, this addon requires R2!
More info at Intel RealSense SDK page

To create new project:

  • after installing Intel RealSense SDK 2016 R2, clone this github repo into your ofx/addons directory
  • use oF project generator, include ofxRealSense addon, open project in VS2015
  • right click on project in Solution Explorer, select Properties
  • change "Configuration" to "All Configurations" and "Platform" to "All Platforms"
  • under "C\C++" > "General" > "Additional Include Directories", add:
    $(RSSDK_DIR)include;
  • under "Linker" > "General" > Additional Library Directories", add:
    $(RSSDK_DIR)lib$(Platform);

example:
additional include and library directories

###Version Notes:

This addon is based on Seth Gibson's ofxRSSDKv2, but with a re-structured interface and design.

####Features:

In Progress/Working

  • RGB & Depth camera --> ofPixels & ofTexture
  • Image mapping from rgb<-->depth
  • Face Scanning module: preview, scan, export to OBJ, MTL, JPG, JSON (face tracking landmarks)

To Do:

  • threaded synced RGB & Depth capture
  • auto point cloud (auto-mesh?)
  • face tracking module
  • face expression module
  • person (skeleton) tracking module

###Stay Tuned!

Pull requests welcome.

You can’t perform that action at this time.