Google Inc.

Palo Alto Networks, Inc.

Hourglass Technologies, Inc.

Honeywell International Inc.

Wing Aviation LLC

Fan Zhang

Jason Gouliard

Amit Ghorawat

Jennifer Cordova

Madeline Chan

Zhen Wang

Jawwad Wasti

Dylan Taffe

Sharon Lee

Peng Zhuang

Sada Kurapati

Matthias Grundmann

Sheng Ma

Darin Howard Buxbaum

Yi Meng

Valentin Bazarevsky

Andrei Vakunov

Charles S Hsu

Andrei Tkachenka

Ivan T Tzvetanov

Jennifer T Blundo

Feng Quan Liu

Ke Tian

Gabrielle Zadina

Ran Xia

Jialiang Zhang

Gary Kim

Ming Lei

Kejia Shao

Gary Hyun Jin Kim

Dan Towery

Sarah Masters

Joshy Joseph

Shiba Sheikh

Diego Rivas

Ryosuke Matsumoto

Jesse Johnston

Min Kim

Nupur Jain

Nicolas H Remy

Anne-Claire Haury

Alexander James Rickabaugh

Ahmed Fakhry

Xiyuan Xia

Example aspects of the present disclosure are directed to computing systems and methods for hand tracking using a machine-learned system for palm detection and key-point localization of hand landmarks. In particular, example aspects of the present disclosure are directed to a multi-model hand tracking system that performs both palm detection and hand landmark detection. Given a sequence of image frames, for example, the hand tracking system can detect one or more palms depicted in each image frame. For each palm detected within an image frame, the machine-learned system can determine a plurality of hand landmark positions of a hand associated with the palm. The system can perform key-point localization to determine precise three-dimensional coordinates for the hand landmark positions. In this manner, the machine-learned system can accurately track a hand depicted in the sequence of images using the precise three-dimensional coordinates for the hand landmark positions.

Scalable real-time hand tracking

Display screen or portion thereof with transitional graphical user interface

Display screen or portion thereof with graphical user interface

Identifying Internet of Things (IoT) devices with packet flow behavior including by using machine learning models is disclosed. Information associated with a network communication of an IoT device is received. A determination of whether the IoT device has previously been classified is made. In response to determining that the IoT device has not previously been classified, a determination is made that a probability match for the IoT device against a behavior signature exceeds a threshold. Based at least in part on the probability match, a classification of the IoT device is provided to a security appliance configured to apply a policy to the IoT device.

IoT device identification with packet flow behavior machine learning model

Aspects of the technology utilize an integrated computing device that supports a shared, communal interaction while still enabling a personalized experience. This includes a unified desktop experience based on presence, a shared family or common space as part of an ambient visual display, and seamless switching involving user interface elements when the display device is rotatable. A presence sensor detects whether someone is at or near the device. When at least one person is identified, specific information is surfaced on a display based on the current context, enabling the person(s) to interact with the device in selected ways. The device evaluates a current presentation state of the display and selects content items to present. The presentation state can include content currently being displayed, other context signals, or any other information associated with the state of the device, such as whether the device is locked or logged-in, for example.

Computer platform with shared user experience interface

A cardinality query associated with a specific attribute is received. One or more in-scope attribute cardinality partitions is enforced on session record analytics. Vertical data compression is performed to eliminate a time dimension. Horizontal data compression is performed to eliminate one or more out-of-scope attributes. One or more like in-scope session records is aggregated. Magnitudes of each in-scope attribute of each enforced cardinality partition are summed. The sum is returned as a response to the cardinality query.

Distributed cardinality optimization

In some embodiments, a computer-implemented method of managing a fleet of unmanned aerial vehicles (UAVs) is provided. A fleet management computing system receives telemetry information from a plurality of UAVs. The fleet management computing system generates a map interface having a plurality of UAV icons based on the telemetry information. The fleet management computing system receives a selection of an initial group of UAV icons via the map interface, wherein the initial group of UAV icons includes two or more UAV icons. The fleet management computing system receives a de-selection of one or more UAV icons from the initial group of UAV icons to create a final selected group of UAV icons. The fleet management computing system transmits a command to UAVs associated with the UAV icons of the final selected group of UAV icons.

Systems and methods for concurrent management of multiple unmanned aircraft

Growing community of inventors

Mountain View, CA, United States of America

Fan Zhang