Microsoft Technology Licensing, LLC

Other

Andrew William Fitzgibbon

Jamie Daniel Joseph Shotton

Shahram Izadi

Pushmeet Kohli

Stephen Edward Hodges

David Kim

Otmar Hilliges

David G Molyneaux

Toby Leonard Sharp

Antonio Criminisi

Thomas Joseph Cashman

David Alexander Butler

Richard Malcolm Banks

Duncan Paul Robertson

Richard Newcombe

Cem Keskin

Jamie Shotton

Federica Bogo

Patrick Luke Olivier

Matthew Darius Cook

Jonathan James Taylor

Darren Keith Edge

Samuel Gavin Smyth

Jonathan Taylor

Advait Sarkar

Erroll William Wood

Michael William Armstrong

Siân Elizabeth Lindley

Ewa Luger

Olivia Evelyn Mary Fletcher Vane

Benjamin Michael Glocker

Ryan M Geiss

Andrew Blake

Matthew L Bronder

Reinhard Sebastian Bernhard Nowozin

Stephanie Horn

Oliver M Williams

Christoph Rhemann

Christopher Zach

Lewey Geselowitz

Richard E Moore

Ankur Agarwal

Ross Brook Girshick

Luca Ballan

Jan Stuehmer

Christopher Douglas Edmonds

Nicolas Villar

Mark James Finocchio

Andrew Clark Goris

Christian Holz

Haiyan Zhang

Abigail Jane Sellen

William A S Buxton

Nicholas Yen-Cherng Chen

Carsten Rother

Greg Saul

Brian L Hastings

Pashmina Cameron

David Sweeney

Bruce Justin Lindbloom

Lucas Bordeaux

Alexander Rav-Acha

Daniel Stefan Tarlow

Jingjing Shen

Jeremy Martin Jancsary

Martin Szummer

Daniel Cletheroe

Vuk Jovanovic

Vladimir Carapic

Filip Panjevic

Taras Khapko

Jelena Mojasevic

Mihai-Dan Budiu

Robert Corish

Jiawen Chen

Mat Cook

Paul Malcolm McIlroy

Oscar Salandin

Abner Guzmán-Rivera

Danyal Khashabi

Lorenzo Torresani

Mark Finnochio

Bill Buxton

Srikumar Ramalingam

Mohammad Sadegh Ali Akbarian

Eduardo Alberto Soto

Nikola Milosavljevic

David Joseph New Tan

Sameh Khamis

A method for predicting the pose of an articulated object includes receiving spatial information for n joints of the articulated object. The spatial information for the n joints is passed to a machine learning model previously trained to receive spatial information for n+m joints as input, wherein m>=1. From the machine learning model, a pose prediction for the articulated object is received as output based at least on the spatial information for the n joints, and without spatial information for the m joints.

Pose prediction for articulated object

A control system comprises a memory storing a sequence of sensor data received from one or more sensors. The control system has a processor which processes the sensor data to compute a sequence of derived sensor data values. An autoencoder receives the sequence of derived sensor data values and computes a forward prediction of the sequence of derived sensor data values, the autoencoder having been trained imposing a relationship on positions of the derived sensor data values encoded in a latent space of the autoencoder. A processor initiates control of an apparatus using the forward prediction.

Control system using autoencoder

Camera or object pose calculation is described, for example, to relocalize a mobile camera (such as on a smart phone) in a known environment or to compute the pose of an object moving relative to a fixed camera. The pose information is useful for robotics, augmented reality, navigation and other applications. In various embodiments where camera pose is calculated, a trained machine learning system associates image elements from an image of a scene, with points in the scene's 3D world coordinate frame. In examples where the camera is fixed and the pose of an object is to be calculated, the trained machine learning system associates image elements from an image of the object with points in an object coordinate frame. In examples, the image elements may be noisy and incomplete and a pose inference engine calculates an accurate estimate of the pose.

Camera/object pose from predicted coordinates

Examples are disclosed that relate to a camera model for a machine vision application. One example provides instructions executable to receive image data obtained by an image sensor of a camera, the image data capturing a calibration pattern comprising a plurality of calibration features, for each of one or more imaged calibration features in the image data, determine an object space location of the imaged calibration feature, and determine a distance between the object space location and a corresponding ray of a camera model, the camera model defining a plurality of rays that each represent a relationship of an image space location on the image sensor to object space. The instructions are further executable to determine a value of a cost function based on the distances, adjust the camera model until the cost function meets a target condition, and use the camera model in a machine vision application.

Calibrating a machine vision camera

A set of electronic slide decks is searched using a target slide of a first electronic slide deck, each of the electronic slide decks containing a sequence of slides to be presented individually. A computer system receives from a user input individually selecting the target slide as a search input from the slides of the first electronic slide deck, compares the target slide with the slides of each electronic slide deck in the set to determine whether any of those slides matches the target slide, so as to determine, for each of the set of electronic slide decks, whether that electronic slide deck contains at least one slide that matches the target slide; and displays a set of search results in response to the user input, wherein each of the search results represents one of the electronic slide decks determined to contain at least one matching slide.

Managing electronic slide decks

There is an apparatus for detecting pose of an object. The apparatus comprises a processor configured to receive captured sensor data depicting the object. It also has a memory storing a parameterized model of a class of 3D shape of which the object is a member, where an instance of the model is given as a mapping from a point in a 2D rectangular grid to a 3D position. The processor is configured to compute values of the parameters of the model by calculating an optimization to fit the model to the captured sensor data, using the parametrized mapping. The processor is configured to output the computed values of the parameters comprising at least global position and global orientation of the object.

Detecting pose of 3D objects using a geometry image

In various examples there is an apparatus for detecting position and orientation of an object. The apparatus comprises a memory storing at least one frame of captured sensor data depicting the object. The apparatus also comprises a trained machine learning system configured to receive the frame of the sensor data and to compute a plurality of two dimensional positions in the frame. Each predicted two dimensional position is a position of sensor data in the frame depicting a keypoint, where a keypoint is a pre-specified 3D position relative to the object. At least one of the keypoints is a floating keypoint depicting a pre-specified position relative to the object, lying inside or outside the object's surface. The apparatus comprises a pose detector which computes the three dimensional position and orientation of the object using the predicted two dimensional positions and outputs the computed three dimensional position and orientation.

Detecting pose using floating keypoint(s)

A system is provided that allows a user to visualize data. A dataset that includes a plurality of data items arranged in a two-dimensional (2D) format is received. A request to visualize the dataset in three dimensions is then received. A three-dimensional (3D) visualization of the dataset is then generated based on this request. This 3D visualization adds a third dimension to the 2D arrangement of data items, where the extent of the third dimension is commensurate with the type and value of the data items. This 3D visualization includes an initial depiction of the dataset which is generated from a point of view that is specified by this request. The initial depiction of the dataset is then displayed on a display device of the system in lieu of the dataset itself.

Using perspective to visualize data

An apparatus for detecting pose of an object is described. The apparatus has a processor configured to receive captured sensor data depicting the object. The apparatus has a memory storing a model of a class of object of which the depicted object is a member, the model comprising a plurality of parameters specifying the pose, comprising global position and global orientation, of the model. The processor is configured to compute values of the parameters of the model by calculating an optimization to fit the model to the captured sensor data, wherein the optimization comprises iterated computation of updates to the values of the parameters and updates to values of variables representing correspondences between the captured sensor data and the model, the updates being interdependent in computation. The processor is configured to discard updates to values of the variables representing correspondences without applying the updates.

Method and apparatus for prefetching data items to a cache

This disclosure relates to computer-implemented management of first and second electronic documents, each of containing a sequence of content components. A user interface displays a first sequence of display elements to represent the first electronic document, and a second sequence of display elements to represent the second electronic document. Each of the display elements in each of the displayed sequences represents an individual content component or group of content components of the electronic document represented by that sequence. The first and second sequences are displayed alongside each other, with their respective display elements in an initial alignment, to allow a side-by-side comparison of the electronic documents. The user can automatically realign the sequences to a selected content component.

Managing electronic documents

In one or more implementations, a static geometry model is generated, from one or more images of a physical environment captured using a camera, using one or more static objects to model corresponding one or more objects in the physical environment. Interaction of a dynamic object with at least one of the static objects is identified by analyzing at least one image and a gesture is recognized from the identified interaction of the dynamic object with the at least one of the static objects to initiate an operation of the computing device.

Gesture recognition techniques

This disclose relates to computer-implemented management of an electronic slide deck, the electronic slide deck containing a set of slides to be presented individually. The computer system processes the slides of the electronic slide deck to autonomously group the slides into a plurality of groups, the autonomous grouping being performed by identifying the slides in each of the groups as having related content, generates in association with the electronic slide deck, grouping data in electronic storage accessible to the computer system, the grouping data denoting the identified groups of slides, and controls a user interface based on the stored grouping data to convey, to a user, information relating to at least one of the identified groups of slides.

A flexible multi-layer sensing surface comprises a first flexible layer and a second flexible layer. The first flexible layer is a sensing layer and the second sensing layer is another sensing layer or an output layer. The sensing surface also comprises a calibration module which is configured to use the first flexible sensing layer to detect the relative position of the first and second flexible layers and this detected relative position is then used to update one or more stored operating parameters.

Flexible multi-layer sensing surface

Growing community of inventors

Cambridge, United Kingdom

Andrew William Fitzgibbon