Microsoft Technology Licensing, LLC

Craig C Peeper

Tommer Leyvand

Szymon Piotr Stachniak

Johnny Hsienchow Lee

Johnny Chung Lee

Shao Liu

Robert Matthew Craig

Momin Mahmoud Al-Ghosien

Ryan M Geiss

Jamie Daniel Joseph Shotton

Mark James Finocchio

Oliver M Williams

Matt Bronder

Amar Dinesh Patel

David Floyd Aronson

Simon Stachniak

Loren McQuade

Peter-Pike Johannes Sloan

Samuel Z Glassenberg

Daniel Kurt Baker

Charles Claudius Marais

Casey Leon Meekhof

Anuj Bharat Gosalia

Ketan K Dalal

Andrew L Bliss

Vladimir Tankovich

John Lee Rapp

Michael Vincent Oneppo

Patrick Orville Cook

Mark M Lacey

Joseph C Bertolami

Bhaven P Dedhia

Nazeeh Amin Eldirghami

Parham Mohadjer

Spencer Dean Perreault

Jeffrey M J Noyle

Jon D Pulsipher

James Chadon Foster

Michael Taylor Janney

Kien Toan Huynh

Robert John Fitzgerald

Wayne Richard Bisson

Tommy Leyvand

Sam Glassenberg

Goetz Graefe

Daniel Reed Baker

Mark Stefan Grossman

Charles Neil Boyd

Cesar Alejandro Galindo-Legaria

Michael A Toelle

Nigel Robin Ellis

Ajay Kalhan

Zsolt Mathe

Stanley William Adermann

Peter Alan Carlin

Michael C Johnson

Mark Plagge

Emad Barsoum

Brian T Klamik

Iouri Tarassov

Samuel H Smith

Allison Klein

Ronald Forbes

Joe Bertolami

Ronald Omega Forbes, Jr

David C Kline

Tommer Leyand

Samual Z Glassenberg

Channing Pierce Verbeck

Anantha R Kancheria

Harold R Berenson

Anton Rakovchuk

John L Rapp

A candidate human head is found in depth video using a head detector. A head region of light intensity video is spatially resolved with a three-dimensional location of the candidate human head in the depth video. Facial recognition is performed on the head region of the light intensity video using a face recognizer.

Identification using depth-based head-detection data

Methods and computer-storage media are provided for rendering three-dimensional (3D) graphics by tessellating objects using novel structures and algorithms. Rendering utilizing 'patches,' configurable functions that include a specified number of control points, allows for computation on a per-patch or per-control-point basis, in addition to traditional per-vertex, per-primitive, and per-pixel methods. This produces a number of advantages over previous tessellation methods, including the reuse of computations across existing vertices and the ability to process at a lower frequency. The operations to compute points are simplified in order to optimize system resources used in the process. Transitions from un-tessellated to tessellated objects are smoother utilizing the present invention, while developers have more flexibility in the level of detail present at different edges of the same patch. Detail within a displacement map also can be increased without negative effects associated with previous systems and methods.

Efficient geometric tessellation and displacement

An image such as a depth image of a scene may be received, observed, or captured by a device. A grid of voxels may then be generated based on the depth image such that the depth image may be downsampled. A background included in the grid of voxels may then be discarded to isolate one or more voxels associated with a foreground object such as a human target and the isolated voxels associated with the foreground object may be processed.

Systems and methods for removing a background of an image

An image such as a depth image of a scene may be received, observed, or captured by a device. A grid of voxels may then be generated based on the depth image such that the depth image may be downsampled. A background included in the grid of voxels may also be removed to isolate one or more voxels associated with a foreground object such as a human target. A location or position of one or more extremities of the isolated human target may then be determined.

Methods and systems for determining and tracking extremities of a target

Allocation of memory registers for shaders by a processor is described herein. For each shader, registers are allocated based on the shader's level of complexity. Simpler shader instances are restricted to a smaller number of memory registers. More complex shader instances are allotted more registers. To do so, developers' high level shading level (HLSL) language includes template classes of shaders that can later be replaced by complex or simple versions of the shader. The HLSL is converted to bytecode that can be used to rasterize pixels on a computing device.

Dynamic subroutine linkage optimizing shader performance

An image such as a depth image of a scene may be received, observed, or captured by a device. A grid of voxels may then be generated based on the depth image such that the depth image may be downsampled. A background included in the grid of voxels may also be removed to isolate one or more voxels associated with a foreground object such as a human target. A location or position of one or more extremities of the isolated human target may be determined and a model may be adjusted based on the location or position of the one or more extremities.

Human tracking system

A computer system comprises a hardware interface, a computer-memory device, an update engine, and a posture-recognition engine. The hardware interface is configured to receive depth video of an environment from a camera. The computer-memory device stores a background model of the environment preservable over a reboot of the computer system, the background model including a plurality of trusted coordinates derived from the depth video. The update engine is configured to update the background model, including moving a trusted coordinate to greater depth if an observed pixel is behind the trusted coordinate over a first duration, but retaining the trusted coordinate if the observed pixel depth is in front of the trusted coordinate over the first duration. The posture-recognition engine is configured to recognize posture of a user in front of a background portion of the video, which is bounded by the trusted coordinates of the background model.

Background model for user recognition

A method for controlling a computer system includes acquiring video of a subject, and obtaining from the video a time-resolved sequence of depth maps. An area targeting motion is selected from each depth map in the sequence. Then, a section of the depth map bounded by the area and lying in front of a plane is selected. This section of the depth map is used for fitting a geometric model of the subject.

Three-dimensional foreground selection for vision system

A scene is illuminated with modulated illumination light that reflects from surfaces in the scene as modulated reflection light. Each of a plurality of pixels of a depth camera receive the modulated reflection light and observe a phase difference between the modulated illumination light and the modulated reflection light. For each of the plurality of pixels, an edginess of that pixel is recognized, and the phase difference of that pixel is smoothed as a function of the edginess of that pixel.

Edge preserving depth filtering

An image such as a depth image of a scene may be received, observed, or captured by a device. A grid of voxels may then be generated based on the depth image such that the depth image may be downsampled. A model may be adjusted based on a location or position of one or more extremities estimated or determined for a human target in the grid of voxels. The model may also be adjusted based on a default location or position of the model in a default pose such as a T-pose, a DaVinci pose, and/or a natural pose.

Systems and methods for tracking a model

Growing community of inventors

Kirland, WA, United States of America

Craig C Peeper