multisensory depth-camera experiment
build in openFrameworks in C++ portions based upon openframeworks ofxKinect examples and meshRecorder example from weLoveCode.
including integration of ofxKinect2 designed around a vision of a volumetric stills camera/apparatus for street photography.
(with a nod of the head to Vilém Flusser https://en.wikipedia.org/wiki/Vilém_Flusser)
built under http://openFrameworks.cc v0.9.8 with dependencies to standard addons ofxKinect ofxXmlSetting, and additional addons ofxImGui and ofxCv.
Currently integrated with Kinect v1 (1414) running under OSX, the source includes project files for Xcode v 8.8.3 development environment. including integration of Kinect 2 (model XX).
The software reads kinect data to a variable size 3d mesh with point cloud, faces and wireframe with and without surface normals and illumination. Depth and RGB data can be captured to disk as PNG sequences with EXIF meta data and replayed and manipulated. Basic OpenCV routines are integrated allowing gaussian blur, eroding and dilation of either RGB or Depth (mesh) data on both live and pre-recorded data.
Please refer to the master branch as the most stable version and the project page for details of work in progress, the wiki for more detailed information, controls & Gui, references etc. Please add any issues, bugs and requests to the issues track.
see the wiki for keyboard controls, gui details and related useage information. https://github.com/danbz/volume-camera/wiki
see more about the project and development of custom hardware etc at http://buzzo.com/building-a-volumetric-camera/
Dan Buzzo 2017
"ostranenie (Russian verb): make an audience see common things in an unfamiliar or strange way in order to enhance perception of the familiar."
example screencapture video on vimeo https://player.vimeo.com/video/229040967