Image analysis – Applications – 3-d or stereo imaging analysis
Reexamination Certificate
2000-06-30
2004-09-07
Ahmed, Samir (Department: 2623)
Image analysis
Applications
3-d or stereo imaging analysis
C382S103000, C345S419000
Reexamination Certificate
active
06788809
ABSTRACT:
BACKGROUND
1. Field of the Invention
The present invention relates to the field of computer vision and image and gesture recognition. More specifically, this invention relates to a stereo-based approach for gesture recognition.
2. Background
Gestures enrich human communication in a natural and unobtrusive way. Gestures can also enrich human-computer interaction in a similar way. Gestures can provide part of a more natural human computer interface. As in the past, when decreasing cost and increasing computer power enabled development of graphical interfaces controlled by 2D mouse movements, so too today, computers and processors, and other technology are becoming powerful enough to provide a new avenue of human-computer communication and interaction.
Gesture modeling can be classified as either model-based or appearance-based. Model-based approaches for moving articulated objects such as human hands are computationally expensive and difficult to implement. Conversely, appearance-based models need to take account of changing views and topology. Vision-based gesture recognition has been applied successfully to numerous applications such as computer game navigation, TV remote control, American Sign Language recognition, and virtual navigation. The recent availability of stereo camera systems has stimulated new research on stereo-based tracking. Some recent systems use stereo cameras to track self-occluding articulated objects or to combine stereo with other cues to track people. One prior art system presents a method for fitting a complex articulated model of the hand to stereo data. Another prior art system combines range, edge, and color information to track fingers. Image unwarping based on model orientation has been used in the field of face recognition and has, up to now not been applied in the field of hand gesture recognition using stereo images.
Much of the previous work in gesture recognition excludes stereo images. The prior art includes use of gradient orientation histograms for recognizing hand gestures under constrained movement conditions. Another technique combines gradient histograms with a color probability density tracker to allow hand gesture recognition with free movement. And another method uses a downward pointing camera mounted in a baseball cap to aid in color segmentation of the user's hands for American Sign Language recognition.
Visual interpretation of human gestures by computers promises a simple and unobtrusive method of human-computer interaction. Yet, a system that can distinguish human gestures in a variety of settings regardless of environmental interference has been elusive.
REFERENCES:
patent: 6128003 (2000-10-01), Smith et al.
patent: 6363160 (2002-03-01), Bradski et al.
patent: 6394557 (2002-05-01), Bradski
patent: 6538649 (2003-03-01), Bradski et al.
Hongo, H. et al., “Focus of attention for face and hand gesture recognition using multiple cameras”, Automatic Face and Gestu Recognition, 2000. Proceedings. Fourth IEEE International Conference, pp 156-161, Mar. 2000.*
Koichi Ishibunchi et al., “Real Time Hand Shape Recognition Using Pipe-line Image Processor”, IEEE International Workshop o robot and Human Communication, pp. 111-115, 1992.*
Luis Goncalves et al., “Monocular tracking of the human arm in 3D”, Proc. 5th Int. Conf. On Computer Vision, pp. 764-770, IEEE, 1995.*
W.T. Freeman et al., “Computer Vision for computer games”, Int. Comf on Automatic Face and Gesture Recognition, pp 100-105, 1996.*
Chung-Lin Hung et al., “A model-based Complex Background Gesture Recognition System”,IEEE, vol. 1, pp. 93-98, 1996.*
“Parameter estimation techniques: a tutorial with application to conic fitting”, Zhengyou Zhang, Image and Vision Computing 15 (1997) 59-76; 1997 Elsevier Science B.V.
Bouget Jean-Yves
Bradski Gary R.
Chu Michael H.
Grzeszczuk Radek
Ahmed Samir
Yates Steven D.
LandOfFree
System and method for gesture recognition in three... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with System and method for gesture recognition in three..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and System and method for gesture recognition in three... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3205013