Subsribe to our RSS Feed

Natural User Interface

Overview

In computing, a natural user interface, or NUI, is the common parlance used by designers and developers of computer interfaces to refer to a user interface that is effectively invisible, or becomes invisible with successive learned interactions, to its users. The word natural is used because most computer interfaces use artificial control devices whose operation has to be learned. A NUI relies on a user being able to quickly transition from novice to expert. While the interface requires learning, that learning is eased through design which gives the user the feeling that they are instantly and continuously successful. This can be aided by technology which allows users to carry out relatively natural motions, movements or gestures that they quickly discover control the computer application or manipulate the on-screen content. A common misunderstanding of the "Natural User Interface" is that it is somehow mimicry of nature or that some inputs to a computer are somehow more 'natural' than others. In truth, the goal is to make the user feel like a natural.                                                     - from wikipedia -

Current research trends in NUI are human motion analysis and speech analysis. We are interested in 2D and 3D based human motion analysis as a natural user interface. Human motion analysis means producing information based on the apparent human motion in two or more consecutive 2D/3D datum. There are many categories of human motion including hand motion, arm motion, head motion, and gait. Hand gesture recognition for remote control is our primary research topic. In addition, we have interest in rapid face detection for analyzing head motion.

Research

Hand Gesture Recognition

Title Summary Movie
Kinect based Finger Tracking

We have developed the finger tracker using MS Kinect. This tracker recognizes two statuses - one finger status is for drawing, five finger statuses is for clearing. Based on this technology, we are developing hand gesture recognition for remote control.
Hand Gesture Recognition for Remote Control

We are developing hand gesture recognition for remote control based on MS Kinect. This demo shows that the user can control the smart device using hand gestures like shake, push, clockwise circle, counter-clockwise circle, etc.


Rapid Face Detection

Title Summary Movie
Color filtering based efficient face detection

We have developed the efficient face detector based on color filtering. By adopting an efficient scanning and rejection method for the face detector, we can accelerate most kinds of face detectors.