An international conference connecting people
in CAD research, education and business
Bookmark and Share
Copyright (C) CAD Solutions, LLC. All rights reserved.
Proceedings of CAD'14, 2014, 166-168
Generating 3D CAD Art from Human Gestures using Kinect Depth Sensor

Miri Weiss Cohen, Alex Frid, Mark Melin, Vladimir Ladijenski, Braude College of Engineering, Israel

Abstract. Microsoft Kinect was originally designed as a motion-sensing device developed as an Xbox console controller that facilitates interaction using gestures and body motion, but its applicability goes beyond the gaming domain as it can be used in robotics, medicine  and arts [1,5,6]. In this study we propose the generation of visual interactive art with 3D geometric features using Microsoft Kinect sensor. Natural human movement and gesture recognition are used for creation and interaction with various objects in 3D space for art design. The art design results could then be captured to yield a static art form, a dynamic video art design or an interactive reactive video art, all providing another dimension to the art domain. An algorithm has been developed that uses the Kinect output as it records movement to create a structure that will accurately represent the movement captured and to preserve the sequence of movements either for future works or to apply transformations on existing structures. The last stage is to convert the structured data into its final state as art design. The proposed algorithm consists of three main stages (1) time dependent depth data acquisition and joints identification, followed by (2) weighted undirected graph generation that passed to (I3) graph scanning algorithm with visual conversion and post processing.

Keywords. 3D CAD art, kinect sensor, natural user interface (NUI), computer art

DOI: 10.14733/cadconfP.2014.166-168