Instant 3D design concept generation and visualization by real-time hand gesture recognition

Jinsheng Kang, Kang Zhong, Sheng-feng Qin, Hongan Wang, David Wright

Research output: Contribution to journalArticlepeer-review

30 Citations (Scopus)

Abstract

3D conceptual design is a creative process. While the designers are in the inspirational mode, they want to quickly express the design ideas in 3D and visualize it in real time without obstacle. In an exploration of a new, natural and more user friendly interface to assist rapid 3D conceptual generation and visualization, a prototype system of hand gesture and motion capture based user interface has been proposed and implemented. In this paper, we present the framework and components of this real time 3D conceptual generation and visualization system. Hand gestures were designed with the consideration of ease to use and the suitability for real time continuous recognition. Rule-based intelligent menu/icons were designed which changes the prompts according to the design process and expected operations. Real time hand gesture recognition was realized by skeleton model based template matching, and the use of hidden Markov models (HMMs). The recognized hand gestures become command script in OpenSCAD environment, and the 3D design concept was instantly generated and displayed on the screen. A case study was conducted for the initial evaluation of the system.
Original languageEnglish
Pages (from-to)785-797
JournalComputers in Industry
Volume64
Issue number7
DOIs
Publication statusPublished - Sep 2013

Fingerprint

Dive into the research topics of 'Instant 3D design concept generation and visualization by real-time hand gesture recognition'. Together they form a unique fingerprint.

Cite this