Intel Corporation

Ke Han

Ke Ding

YuHuan Huang

Lili Ma

Dong Wang

Xun Wang

Hemin Han

Xiaodong Cai

Zhen Zhou

Jingyi Ma

Liang Li

Stephan Jourdan

Mingqiu Sun

Lu Wang

Prakash N Iyer

Hua He

Lev Yitzhak Lavy

Andrzej Mialkowski

Yehiel Shilo

Shouwei Sun

He Han

Guangyu Ren

Qin Duan

Jesse R Walker

Tao Peng

Arvind A Kumar

Yan Chen

Feng Xu

Liang Cheng

Yu Yu

Rajasekaran Andiappan

Jinkui Ren

Zhiqiang Qin

Xiaoyong Du

Kun-Feng Lin

Guoquan Shen

Qiuhua Hu

Jixing Gu

Dror Lederman

Zhifeng Wang

Renjie Cui

Mengchun Lv

Venkat Raghavulu

Weize Ye

The present disclosure relates to automatic user adaptive triggering, such as triggering of always-on vision. A camera may be activated based on a result of applying a set of sensor features to a logic model. A label for the set of sensor features may be determined based on whether a face is detected or a face is not detected in a first set of images captured by the camera. The labeled sensor data may be stored along with historical data. A user behavior model may be generated using a learning algorithm based on the labeled sensor data. The user model may be incrementally updated based on new training data. The user behavior model may be used for determining whether to the activate the camera based on a particular set of sensor features.

Automatic user adaptive triggering

Methods and apparatus to implement always-on context sensor hubs for processing multiple different types of data inputs are disclosed. An examples apparatus includes a first processor core to implement a host controller, and a second processor core to implement an offload engine. The host controller includes first logic to process sensor data associated with an electronic device when the electronic device is in a low power mode. The host controller is to offload a computational task associated with the sensor data to the offload engine. The offload engine includes second logic to execute the computational task.

Methods and apparatus to implement always-on context sensor hubs for processing multiple different types of data inputs

Methods, apparatus, systems and articles of manufacture are disclosed to dynamically schedule a wake pattern in a computing system. An example apparatus includes a device state controller to determine contexts of a device based on sensor measurements collected at a first time, an associator to associate a state of an operating system with the contexts of the device, the state obtained at a second time after the first time, a training controller to generate a prediction model based on the association, the prediction model to predict a third time when the state of the operating system will be active based on the contexts, and a schedule controller to reduce power consumption of the device by triggering a wake event before the third time, the wake event to prepare the device for exiting an inactive state.

Methods, systems, articles of manufacture, and apparatus to dynamically schedule a wake pattern in a computing system

Methods, apparatus, systems and articles of manufacture are disclosed for display panel human presence detection. An example display device includes a sensor to generate a detection signal based on whether a human is detected, and logic circuitry to generate a control signal to cause an adjustment of a power state of the display device based on the detection signal.

Systems, methods, and apparatus for display panel human presence detection

Particular embodiments described herein provide for an electronic device that includes a first housing, where the first housing includes at least one first housing accelerometer and at least one first housing gyroscope, a second housing, where the second housing includes at least one second housing accelerometer and at least one second housing gyroscope, and a hinge. The hinge rotatably couples the first housing to the second housing, where a hinge angle is determined using data from the first housing accelerometer, the second housing accelerometer, the first housing gyroscope, and the second housing gyroscope if the electronic device is not in a portrait orientation where the hinge is vertical to a ground or not in motion and is determined using data from the first housing gyroscope and the second housing gyroscope if the electronic device is in portrait orientation or is in motion.

Hinge angle detection

Technologies for monitoring a health-risk condition of a user include a virtual reality compute device having one or more near infrared (NIR) sensors. The virtual reality compute device presents a virtual reality (VR) presentation to the user. The virtual reality compute device produces sensor data through the one or more NIR sensors that is indicative of a heart rate of the user and a blood pressure of the user while the VR presentation is presented to the user. The virtual reality compute device determines whether the user is in a health-risk condition based on a comparison of the heart rate of the user to a heart rate safety threshold and a comparison of the blood pressure of the user to a blood pressure safety threshold. The virtual reality compute device performs a health-risk condition response in response to a determination that the user is in the health-risk condition.

Technologies for monitoring health-risk condition of user

Technologies for determining a user's location include a mobile computing device to determine, based on sensed inertial characteristics of the device, a walking gait of a user. The walking gait is one of a first gait indicative of the user holding the g device to the user's side or a second gait indicative of the user swinging the device along the user's side. The device further detects that the user has taken a physical step based on the inertial characteristics and the determined walking gait of the user, and determines a raw directional heading of the device indicative of a direction of the physical step. The device determines an estimated location of the user based on the determined raw directional heading, an estimated step length, and the user's previous location.

Technologies for pedestrian dead reckoning

Geofence crossing-based control systems and techniques are described herein. For example, a geofence crossing control technique may include receiving a location signal indicative of a range of locations in which a mobile computing device is located; receiving a velocity signal indicative of a speed and direction of the mobile computing device; generating, for each of a plurality of candidate geofence crossing times, a performance indicator based on the location signal, the velocity signal, and a boundary of the geofence; selecting a geofence crossing time from the plurality of candidate geofence crossing times based on the performance indicators; and transmitting a control signal representative of the geofence crossing time. Other embodiments may be disclosed and/or claimed.

Systems and techniques for geofence crossing-based control

An activity monitoring circuit enables power conservation for an activity monitoring system. The activity monitoring system includes a sensor device that generates activity data based on detected movement. An activity monitoring circuit receives the activity data as input from the sensor device, and processes the activity data. Based on the processing, the activity monitoring circuit signals the sensor device to change a duty cycle of its operation, by adjusting how long it is on and off. The activity monitoring circuit can include a sensor hub that determines an activity state indicated by the activity data and generates a prediction of stability of the determined activity state based on historical activity data. Based on activity state and stability, the sensor hub can determine a duty cycle for operation of the sensor device.

Low power activity monitoring with adaptive duty cycle in a sensor hub

Techniques are provided for unification of classifier models across device platforms of varying form factors and/or sensor calibrations. A methodology implementing the techniques according to an embodiment includes extracting classification features from data provided by sensors associated with a first device platform. The method also includes applying a feature mapping function to the extracted features. The feature mapping function is configured to transform the features such that the are suitable for use by a classifier model that is trained on data provided by sensors associated with a second device platform. The method further includes executing the classifier model on the transformed features to generate classifications, for example recognized activities associated with use of the first device. The feature mapping function is based on application of a statistical distribution distance minimization between a sampling of data provided by sensors of the first device and sensors of the second device.

Unification of classifier models across device platforms

Systems, apparatuses and methods may provide for detecting one or more trigger condition on a mobile platform and activating an offline calibration of one or more sensors on the mobile platform in response to the one or more trigger conditions. Additionally, an operational profile of the one or more sensors may be updated based on the offline condition. In one example, the trigger conditions include one or more of a time condition, a temperature condition, a motion condition or a system event condition.

Offline sensor calibration

A virtual reality headset system includes a vision correction module. The vision correction system can detect a degree of myopia or other visual ailment of a user of the VR system, and then adjust a vision correction lens to adjust for user myopia. An implementation of the vision correction module can adjust right and left lenses separately.

Virtual reality (VR) system with nearsightedness optometry adjustment

Systems and methods may provide for obtaining first sensor data associated with a gyroscope and obtaining second sensor data associated with a magnetometer. Additionally, the first sensor data, the second sensor data and an extended Kalman filter may be used to calibrate the magnetometer. In one example, a sampling rate of the magnetometer is increased before obtaining the second sensor data and the sampling rate of the magnetometer is decreased after calibration of the magnetometer.

Extended Kalman filter based autonomous magnetometer calibration

Techniques are disclosed for enabling an integrated sensor hub of a main computer to access a detachable peripheral device. In an embodiment, a system includes a main unit having a peripheral interface, an embedded controller, and a device controller. The peripheral interface is configured to be detachably coupled to a peripheral. The peripheral includes a control unit and an input/output device. The embedded controller is configured to communicate with the control unit of the peripheral via the peripheral interface while the peripheral is attached to the peripheral interface. The embedded controller includes at least one data register, and in some embodiments, a set of data registers, configured to store data relating to the peripheral and to the corresponding input/output device. The device controller is configured to read data from the data register(s) of the embedded controller, write data to the data register(s) of the embedded controller, or both.

Accessing input/output devices of detachable peripheral by a main computer

Growing community of inventors

Shanghai, China

Ke Han