Facebook, Inc.

Pricewaterhousecoopers Llp

Blackboard Inc.

Benjamin Grady Cunningham

Brian Daniel Amerige

Joseph G Harrington

Jacob T Wilson

Jayakumar Pillai

Abhishek Sanghavi

Lindsey P Lewis

Vinston Sundara Pandiyan Sigamani

Michael James Matas

Shannon M Ma

Methods and systems for artificial intelligence (AI)-assisted document annotation and training of machine learning-based models for document data extraction are described. The methods and systems described herein take advantage of a continuous machine learning approach to create document processing pipelines that provide accurate and efficient data extraction from documents that include structured text, semi-structured text, unstructured text, or any combination thereof.

Continuous machine learning method and system for information extraction

Methods and systems for artificial intelligence-assisted document annotation

Systems, methods, and non-transitory computer-readable media can receive a first user interaction associated with a first transition in a graphical user interface. The first transition is executed in the graphical user interface. A second user interaction associated with a second transition in the graphical user interface is received during the executing the first transition. The second transition is executed in the graphical user interface during the executing the first transition.

Systems and methods for concurrent graphical user interface transitions

Systems, methods, and non-transitory computer-readable media can present a content feed through a display screen, the content feed including one or more content items. At least one content item in the content feed for which a layout is to be generated can be determined. The layout for the content item can be generated prior to the content item being accessed.

Systems and methods for preloading content

A method is performed at an electronic device that includes a display, a touch-sensitive surface, one or more processors, and memory storing one or more programs. The device displays a user interface of a software application, wherein the user interface includes a plurality of user-interface elements. A first gesture is detected on the touch-sensitive surface while displaying the first user interface, and an initial direction of movement is determined for the first gesture. The device recognizes that the initial direction corresponds to one of a first predefined direction on the touch-sensitive surface or a second predefined direction on the touch-sensitive surface, wherein the first predefined direction is distinct from the second predefined direction. Display of one or more user-interface elements of the plurality of user-interface elements is manipulated in accordance with the corresponding one of the first or second predefined direction.

Touch-based gesture recognition and application navigation

An electronic device includes a display, a touch-sensitive surface, one or more processors, and memory storing one or more programs. The device displays a first user interface of a hierarchy of user interfaces of a software application associated with first and second pan gesture recognizers. The first pan gesture recognizer is configured to recognize a pan gesture that has an initial direction along a first axis and/or a first direction. The second pan gesture recognizer is configured to recognize a pan gesture that has an initial direction along a second axis and/or a second direction distinct from the first direction. The device detects a first pan gesture in an initial direction across the touch-sensitive surface while displaying the first user interface. The device identifies a pan gesture recognizer configured to recognize the first pan gesture, and processes the first pan gesture using the identified pan gesture recognizer.

In one embodiment, a method includes determining, by a computing device, a scroll range for an image based on a dimension of the image, determining the image's origin position device's initial tilt, determining a second tilt and a rate of rotation about a first axis, and determining a rate of rotation about a second axis. The method includes comparing the device's rate of rotation about the first axis with the device's rate of rotation about the second axis or one of its components. The method further includes determining, based on the comparison, whether to render a scrolling presentation of the image. When the scrolling presentation of the image is to be rendered, the method includes determining a progress parameter based on the scroll range, the origin position, the initial tilt, and the second tilt; and rendering a scrolling presentation of the image based on the progress parameter.

Tilting to scroll

Methods for presenting an orientation aware application on a device are provided. In one aspect, a method includes receiving visual data providing an image of a physical three-dimensional environment and orientation data indicating an orientation of an actual device within the physical three-dimensional environment, and providing, for display on the actual device, a visual representation of a virtual device displaying an orientation aware application using the image of the physical three-dimensional environment. The visual representation of the virtual device appears within a simulated three-dimensional environment that is based on the orientation data indicating the orientation of the actual device within the physical three-dimensional environment. At least a portion of the area surrounding the visual representation of the virtual device includes a display of the simulated three-dimensional environment. Systems and machine-readable storage media are also provided.

Growing community of inventors

New York, NY, United States of America

Benjamin Grady Cunningham