Systems and methods of camera-based fingertip tracking
AssigneeUniversity of Central Florida Research Foundation, Inc.
InventorsKiminobu Sugaya, Andrew Sugaya
Life of this patent
Estimating the term…
Abstract
Systems and methods for camera-based fingertip tracking are disclosed. One such method includes identifying at least one location of a fingertip in at least one of the video frames, and mapping the location to a user input based on the location of the fingertip relative to a virtual user input device.
Claims19
A method of manipulating a virtual object in a series of video frames, the method comprising:
- identifying at least one location of a body part in at least one of the video frames;
- and mapping the location to the virtual object based on the location of the body part relative to the virtual object, wherein said identifying at least one location of a body part comprises identifying a target pixel in a first one of the video frames, the target pixel representing a potential body part location;
- tracking motion of the target pixel in video frames occurring after the first one;
- determining whether the target pixel is a confirmed pixel, the confirmed pixel representing a probable body part location;
- and tracking motion of the confirmed pixel.
A method of translating sign language, the method comprising
- receiving a series of images from a camera pertaining to a motion of a provider's hand, finger, fingertip, and/or arm corresponding to a first unit of sign language information;
- processing the images to correlate the motion to a stored unit of information stored on a computer readable medium, wherein the processing comprises identifying a target pixel in a first on of the series of images representing a potential location of provider's hand, finger, fingertip and/or arm;
- tracking motion of the target pixel in at least one image occurring after the first one;
- determining whether the target pixel is a confirmed pixel representing a probable hand, finger, fingertip and/or arm location, and tracking motion of the confirmed pixel;
- and
- outputting an output unit of information to a receiver in a form different relative to the first unit of information.
A system for manipulating a virtual object, the system comprising an eyewear device comprising a lens that generates an image field that displays a virtual object;
- a camera that monitor's motion in said image field, and a computer programmed to track motion of a user in the image field and mapping the motion to said virtual object, wherein the computer receives a series of images from said camera and is programmed to identify at least one location of a body part in at least one of the images;
- and mapping the location to the virtual object based on the location of the body part relative to the virtual object, wherein said identifying at least one location of a body part comprises identifying a target pixel in a first one of the images, the target pixel representing a potential body part location;
- tracking motion of the target pixel in at least one image occurring after the first one;
- determining whether the target pixel is a confirmed pixel, the confirmed pixel representing a probable body part location;
- and tracking motion of the confirmed pixel.
A system for translating sign language, the system comprising a camera in communication with a computer, the computer comprising a computer readable media and being programmed to receive video image of a unit of sign language motion, correlating the sign language motion to unit of information stored on the computer readable media;
- and an output device for outputting an output unit of information, wherein the computer is programmed to output the output unit of information via said output device, wherein the camera provides a series of images to the computer and the computer is programmed to identify a target pixel in a first on of the series of images representing a potential location of provider's hand, finger, fingertip and/or arm;
- tracking motion of the target pixel in at least one image occurring after the first one determining whether the target pixel is a confirmed pixel representing a probable hand, finger, fingertip and/or arm location, and tracking motion of the confirmed pixel.
Dependencies are inferred from the claim text.