Apple Inc.

Freedom Scientific, Inc.

Ai Squared

Google Inc.

Carnegie Mellon University

University of Rochester

Jeffrey P Bigham

Aaron M Leventhal

Qi Shan

Samuel C White

Carlos E Guestrin

Mingzhe Li

Xiaoyi Zhang

Christopher Harrison

Gierad Laput

Adrien Treuille

Thiago Teixeira

Bo Robert Xiao

Walter S Lasecki

Jason Wiese

Brian Watson

Walter Lasecki

Calibration of eye tracking is improved by collecting additional calibration pairs when user is using apps with eye tracking. A user input component is presented on a display of an electronic device, detecting a dwelling action for user input component, and in response to detecting the dwelling action, obtaining a calibration pair comprising an uncalibrated gaze point and a screen location of the user input component, wherein the uncalibrated gaze point is determined based on an eye pose during the dwelling action. A screen gaze estimation is determine based on the uncalibrated gaze point, and in response to determining that the calibration pair is a valid calibration pair, training a calibration model using the calibration pair.

On-the-fly calibration for improved on-device eye tracking

Techniques for programmatically magnifying one or more visible content elements of at least one markup language document, so as to increase the display size of those visible content elements. A magnification facility may be configured to apply multiple different zoom techniques. The magnification facility may be configured to evaluate the markup language document(s) at a time that the document(s) are being processed for display to select which of the multiple different zoom techniques may be applied at a time to increase a display size of visible content elements relative to a default display size for those elements. The magnification facility may be incorporated within the markup language document(s) and executed by a viewing application that processes markup language documents. For example, the markup language document(s) may form a web page and the magnification facility may be implemented as scripting language code incorporated into the document(s) of the web page.

Techniques for programmatic magnification of visible content elements of markup language documents

The disclosure describes a sensor system that provides end users with intelligent sensing capabilities, and embodies both crowd sourcing and machine learning together. Further, a sporadic crowd assessment is used to ensure continued sensor accuracy when the system is relying on machine learning analysis. This sensor approach requires minimal and non-permanent sensor installation by utilizing any device with a camera as a sensor host, and provides human-centered and actionable sensor output.

System and method for adaptive, rapidly deployable, human-intelligent sensor feeds

Example systems and methods may facilitate processing of voice commands using a hybrid system with automated processing and human guide assistance. An example method includes receiving a speech segment, determining a textual representation of the speech segment, causing one or more guide computing devices to display one or more portions of the textual representation, receiving input data from the one or more guide computing devices that identifies a plurality of chunks of the textual representation, determining an association between the identified chunks of the textual representation and corresponding semantic labels, and determining a digital representation of a task based on the identified chunks of the textual representation and the corresponding semantic labels.

Transcription and tagging system

In a computer system receiving input from a user via at least a keyboard and a pointing device, in which input via the pointing device causes corresponding movement of a pointing image on a display screen of the computer system, user input may be received via the pointing device to point the pointing image at an onscreen object on the display screen. In response to the user activating a key on the keyboard while the pointing image is pointing at the onscreen object, a secondary action with respect to the onscreen object may be executed.

Executing secondary actions with respect to onscreen objects

In a computer system having a display screen configured to display visual content, a plurality of techniques may be identified to be considered for enhancing visual accessibility of a particular collection of visual content to be displayed to an end user on the display screen. For each technique, an algorithm may be applied to compute one or more measures of health of the display of the particular collection of visual content resulting from applying the respective technique to enhance the visual accessibility of the particular collection of visual content. Based at least in part on the computed measures of health, one or more best techniques may be selected and applied to enhance the visual accessibility of the particular collection of visual content. The enhanced particular collection of visual content may be displayed to the end user on the display screen.

Selecting techniques for enhancing visual accessibility based on health of display

Growing community of inventors

Pittsburgh, PA, United States of America

Jeffrey P Bigham