View-based interpretation of real-time optical flow for gesture recognition

Cutler, Ross; Turk, Matthew

doi:10.1109/afgr.1998.670984

Cited by 148 publications

(93 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…This average indicates the translation of the hand. But flow fields are able to express patterns of motion beyond translation and have been shown to be useful in gesture recognition tasks [8]. Fig.…”

Section: Beyond Pointingmentioning

confidence: 99%

See 1 more Smart Citation

FlowMouse: A Computer Vision-Based Pointing and Gesture Input Device

Wilson

Cutrell

2005

Human-Computer Interaction - INTERACT 2005

View full text Add to dashboard Cite

Abstract. We introduce FlowMouse, a computer vision-based pointing device and gesture input system. FlowMouse uses optical flow techniques to model the motion of the hand and a capacitive touch sensor to enable and disable interaction. By using optical flow rather than a more traditional tracking based method, FlowMouse is exceptionally robust, simple in design, and offers opportunities for fluid gesture-based interaction that go well beyond merely emulating pointing devices such as the mouse. We present a Fitts law study examining pointing performance, and discuss applications of the optical flow field for gesture input.

show abstract

Section: Beyond Pointingmentioning

confidence: 99%

“…Simple clustering techniques on the position of flow vectors or on their dominant directions of movement (as in [8]) may be used to detect the motion of two hands under the camera. Various two-handed interactions would thus be enabled (see [6] and [12] for examples).…”

Section: Two Handed Interactionsmentioning

confidence: 99%

FlowMouse: A Computer Vision-Based Pointing and Gesture Input Device

Wilson

Cutrell

2005

Human-Computer Interaction - INTERACT 2005

View full text Add to dashboard Cite

show abstract

“…For certain applications, hand trajectories can be sufficient for gesture classification [3,23]. However, in some domains, knowledge of more detailed hand configuration must be used to disambiguate between different gestures; e.g.,in signed languages.…”

Section: Related Workmentioning

confidence: 99%

“…Such systems typically make restrictive assumptions on the domain: only hands move, the hands are the fastest moving objects in the scene [17, 38,40,3,21,23], hands are skin colored, or they are the only skin-colored objects in the scene [21,34]. Often the background is assumed to be static, and known [21,8].…”

Section: Related Workmentioning

confidence: 99%

3D hand pose reconstruction using specialized mappings

Rosales

Athitsos

Sigal

et al.

Proceedings Eighth IEEE International Conference on Computer Vision. ICCV 2001

116

View full text Add to dashboard Cite

show abstract

“…Computationally simple view-based approaches to action recognition have also been proposed [1,2]. However, these systems do not attempt intentional tracking and modelling to control active camera views for VMI.…”

Section: Introductionmentioning

confidence: 99%