Canesta, Inc.

Microsoft Technology Licensing, LLC

Leland Stanford Junior University

Carlo Tomasi

Salih Burak Gokturk

Abbas Rafii

Fahri Sürücü

David Seungwon Paik

Sandy A Napel

Xinqiao Liu

Cyrus Soli Bamji

Christopher Allen Beaulieu

Burak Acar

Christopher Frederick Beaulieu

R Brooke Jeffrey, Jr

Acar Burak

Fahri S&uuml;r&uuml;c&uuml;

Acquired three-dimensional positional information is used to identify user created gesture(s), which gesture(s) are classified to determine appropriate input(s) to an associated electronic device or devices. Preferably at at least one instance of a time interval, the posture of a portion of a user is recognized, based at least one factor such as shape, position, orientation, velocity. Posture over each of the instance(s) is recognized as a combined gesture. Because acquired information is three-dimensional, two gestures may occur simultaneously.

Gesture recognition system using depth perceptive sensors

A sensor system is provided for determining a deployment level of an airbag in a vehicle. A light source of the sensor system emits light onto a region around a vehicle seat. An array of light-sensitive pixels which capture reflected light from the scene, including reflected light that originated from the light source. Processing resources are provided that determine depth information for an object in the scene based on a time-of-flight characteristic of the reflected light from the light source captured on the array. The processing resources may be configured to determine occupancy data for the object based on the captured reflected light from the scene. The processing resources are configured to determine the deployment level of the airbag based at least in part on the occupancy data in when a collision of the vehicle occurs.

System and method for providing intelligent airbag deployment

A detection and classification method of a shape in a medical image is provided. It is based on generating a plurality of 2-D sections through a 3-D volume in the medical image. In general, there are two steps. The first step is feature estimation to generate shape signatures for candidate volumes containing candidate shapes. The feature estimation method computes descriptors of objects or of their images. The second general step involves classification of these shape signatures for diagnosis. A classifier contains, builds and/or trains a database of descriptors for previously seen shapes, and then maps descriptors of novel images to categories corresponding to previously seen shapes or classes of shapes.

Three-dimensional pattern recognition method to detect shapes in medical images

Three-dimensional position information is used to identify the gesture created by a body part of interest. At one or more instances of an interval, the posture of a body part is recognized, based on the shape of the body part and its position and orientation. The posture of the body part over each of the one or more instances in the interval are recognized as a combined gesture. The gesture is classified for determining an input into a related electronic device.

A class of measurement devices can be made available using a family of projection patterns and image processing and computer vision algorithms. The proposed system involves a camera system, one or more structured light source, or a special pattern that is already drawn on the object under measurement. The camera system uses computer vision and image processing techniques to measure the real length of the projected pattern. The method can be extended to measure the volumes of boxes, or angles on planar surfaces.

Optical methods for remotely measuring objects

A method to detect and classify a structure of interest in a medical image is provided to enable high specificity without sacrificing the sensitivity of detection. The method is based on representing changes in three-dimensional image data with a vector field, characterizing the topology of this vector field and using the characterized topology of the vector field for classification of a structure of interest. The method could be used as a stand-alone method or as a post-processing method to enhance and classify outputs of a high-sensitivity low-specificity method to eliminate false positives.

Method for detecting and classifying a structure of interest in medical images

A projection array is provided that comprises a plurality of discrete projection elements. An image array is obtained of a scene with the light projected onto it that is coded using the projection array. Correspondence information is determined for each element in the image array, where the correspondence information can be used to determine which of the plurality of elements in the projection array corresponds to a particular image element. The determination of correspondence information for each element in the image array can be made independently of correspondence information for other elements in the image array.

Coded-array technique for obtaining depth and other position information of an observed object

Structured light is directed across a monitored region. An image is captured of a light pattern that forms on the object as a result of the object intersecting the structured light when the object is placed at a first position in the monitored region. A geometric characteristic is identified of the image of the light pattern. The geometric characteristic is variable with a depth of the first position relative to where the image is captured. The depth of the first position is approximated based on the measured geometric characteristic.

Method and apparatus for approximating depth of an object's placement onto a monitored region with applications to virtual interface devices

A system used with a virtual device inputs or transfers information to a companion device, and includes two optical systems OS , OS . In a structured-light embodiment, OS emits a fan beam plane of optical energy parallel to and above the virtual device. When a user-object penetrates the beam plane of interest, OS registers the event. Triangulation methods can locate the virtual contact, and transfer user-intended information to the companion system. In a non-structured active light embodiment, OS is preferably a digital camera whose field of view defines the plane of interest, which is illuminated by an active source of optical energy. Preferably the active source, OS , and OS operate synchronously to reduce effects of ambient light. A non-structured passive light embodiment is similar except the source of optical energy is ambient light. A subtraction technique preferably enhances the signal/noise ratio. The companion device may in fact house the present invention.

Quasi-three-dimensional method and apparatus to detect and localize interaction of user-object and virtual transfer device

A plurality of sound-detecting devices are configured to detect sound from an event of interest. Information about the sound is recorded that is dependent on a distance between the source position and the location of the sound-detection devices. The source position is approximated using the recorded information and the relative position of individual sound-detection devices to one another. The approximated source position is used as input for operating a electronic device.

Growing community of inventors

Palo Alto, CA, United States of America

Carlo Tomasi