20-03-2012, 02:59 PM
Vision-Based Gesture Recognition:
10.1.1.42.346.pdf (Size: 140.63 KB / Downloads: 75)
Introduction
The evolution of user interface (UI) witnessed the development from text-based
UI based on keyboard to GUI based on mice. What will be the counterpart of
mouse when we are trying to explore 3D virtual environments (VEs) in Human
Computer Intelligent Interaction (HCII) or Perceptual User Interface (PUI)? In
current VE applications, keyboards, mice, wands and joysticks are still the most
popular and dominant devices. However, they are inconvenient and unnatural.
The use of human movements, especially hand gestures, has become an impor-
tant part of HCII in recent years, which serves as a motivating force for research
in modeling, analyzing and recognition of hand gestures. Many techniques devel-
oped in HCII can be extended to other areas such as surveillance, robot control
and teleconferencing.
Human Gesture Representation
There have been many studies on human gestures in psycholinguistic research.
Stokoe [44] represents gestures as four aspects which are hand shape, position,
orientation and movement. Kendon [26] describes a philology of gesture, which
consists of gesticulation, language-like gestures, pantomimes, emblems, and sign
language. Sign languages are characterized by a specific set of vocabulary and
grammar. Emblems are informal gestural expressions in which the meaning de-
pend on convention, culture and lexicon.
Application Systems
There have been many implemented application systems in many domains such
as virtual environments, smart surveillance, HCII, teleconferencing, sign lan-
guage translation, etc..
Zeller et al. [57] present a visual environment for very large scale biomolecu-
lar modeling application. This system permits interactive modeling of biopoly-
mers by linking a 3D molecular graphics and molecular dynamics simulation
program. Hand gestures serve as the input and controlling device of the vir-
tual environment. Pavlovic and Berry [2] integrate controlling gesture into the
virtual environment BattleField, in which hand gestures are used not only for
navigating the VE, but also as an interactive device to select and move the vir-
tual objects in the BattleField. Ju et al. [25] develop an automatic system for
analyzing and annotating video sequences of technical talks. Speaker’s gestures
such as pointing or writing are automatically tracked and recognized to provide
a rich annotation of the sequence that can be used to access a condensed version
of the talk.