Google Inc.

Telegent Systems, Inc.

Edward Y Chang

Guanfeng Li

Albert Meixner

Ofer Shacham

Daniel Finchelstein

Qingxuan Yang

Jason Rupert Redgrave

William R Mark

Hongji Bao

Dong Zhang

Zhiyu Wang

Donald C Stark

Qiuling Zhu

David Patterson

Timothy James Knight

Shuchang Zhou

Shahriar Rabii

Neeti Desai

Don Stark

Fabrizio Basso

Masumi Reynders

Dingyin Xia

Michelle Tomasko

Szepo Robert Hung

Vlad Constantin Cardei

Faen Zhang

Zoltan Gyongyi

Jian Gong Deng

Yuzhou Zhang

Huacheng Ke

Zhiyuan Liu

Herbert Huang

Andrey Yatsunenko

Li Guanfeng

Xiance Si

Yongqiang Huang

Quingxuan Yang

An image processor unit is described. The image processor unit includes a plurality of inputs to receive at least one input image. The image processor unit includes a plurality of outputs to provide at least one output image. The image processor unit includes a network coupled to the plurality of inputs and the plurality of outputs. The network is to couple at least one of the inputs to at least one of the outputs. The image processor unit includes an image processor circuit coupled to the network. The network to route an input image that is received at one of the inputs to the image processor circuit. The image processor circuit is to execute image signal processing program code to generate a processed output image from the input image. The network is to route the processed output image to at least one of the outputs.

Configurable and programmable image processor unit

A method is described that includes loading an array of content into a two-dimensional shift register. The two-dimensional shift register is coupled to an execution lane array. The method includes repeatedly performing a first sequence including: shifting with the shift register first content residing along a particular row or column into another parallel row or column where second content resides and performing operations with a particular corresponding row or column of the execution lane array on the first and second content. The method also includes repeatedly performing a second sequence including: shifting with the shift register content from a set of first locations along a resultant row or column that is parallel with the rows or columns of the first sequence into a corresponding set of second locations along the resultant row or column. The resultant row or column has values determined from the operations of the first sequence.

Statistics operations on two dimensional image processor

An apparatus is described. The apparatus includes an image processing unit. The image processing unit includes a plurality of stencil processor circuits each comprising an array of execution unit lanes coupled to a two-dimensional shift register array structure to simultaneously process multiple overlapping stencils through execution of program code. The image processing unit includes a plurality of sheet generators respectively coupled between the plurality of stencil processors and the network. The sheet generators are to parse input line groups of image data into input sheets of image data for processing by the stencil processors, and, to form output line groups of image data from output sheets of image data received from the stencil processors. The image processing unit includes a plurality of line buffer units coupled to the network to pass line groups in a direction from producing stencil processors to consuming stencil processors to implement an overall program flow.

Architecture for high performance, power efficient, programmable image processing

Disclosed here are methods and systems that relate to determining a moving direction of a mobile device user. The methods and systems relate to using an inertial navigation system such as an accelerometer and gyroscope to aid in the determination of the moving direction of the user. The methods and systems may receive an acceleration reading associated with the mobile device, and determine a step frequency of the user based on the acceleration reading. The methods and systems may determine a theoretical model to fit the acceleration reading, and may determine the moving direction of the user based on the theoretical model.

Signal processing to extract a pedestrian's moving direction

A system and method for attitude correction is provided. An acceleration and an attitude of an electronic device are detected. A period of time where a velocity of the electronic device at the beginning of the period of time and a velocity of the electronic device at the end of the period of time are equal is identified. An attitude correction is calculated based on the identified period of time and the detected acceleration of the electronic device during the period of time. The detected attitude of the electronic device is corrected with the calculated attitude correction.

System and method for attitude correction

In one implementation, a computer-implemented method includes receiving at a server a request from an advertiser to target an ad to users of a computer-implemented social network, the request comprising data representing characteristics of the users that the advertiser desires to target. The method further includes ranking the users based on how similar the users' characteristics are to the received characteristics and an influence score for each user that indicates how influential the user is within the social network. The method also includes scoring the advertiser's request based on a bid from the advertiser for an opportunity to display the ad to one or more of the users and assigning the opportunity to display the ad to the one or more users based on a correlation between a score of the advertiser's request and one or more rankings of the one or more users.

Bidding on users

A method is described that includes identifying a set of features of an object, the features being tracked in an image captured by a camera. The method also includes creating a field of vectors for the reference points. The vectors indicate magnitude and direction of change in position of the reference points across more than one frame of the image. The method further includes identifying existence of out of plane movement of the object's features from same radial orientation of the vectors. The method further includes determining an amount of closer/farther movement of the object's features to/from the camera from change in distances between a plurality of the reference points. The method further includes adjusting a position of camera's lens in view of the amount of closer/farther movement of the object's features to keep the camera focused on the object.

Optical flow based auto focus

A MEMS type sensor is used for measuring a particular parameter on an IC chip within an electronic device, such as music player, a smart cellular telephone, etc. The parameter may be a scalar parameter, such as a directional orientation indicative a current compass point, or a multidimensional vector parameter, such as a three-dimensional acceleration. The sensor output is recalibrated using stored coefficients when ambient conditions vary. The stored coefficients may be modified during calibration.

Dynamic calibration of MEMS sensors

A computer-implemented technique for feature extraction includes obtaining an electronic image of an object and performing an edge detection algorithm on the electronic image. The technique further includes performing an edge pooling algorithm and sampling the electronic image edge patches, color patches and texture patches. A set of patches is selected from the edge patches, color patches and texture patches by selecting an (i+1) patch to be within the set of patches based on a Euclidean distance from an ipatch of the set of patches for each of the set of edge patches, the set of color patches and the set of texture patches. A part selection algorithm and a part pooling algorithm is performed to obtain parts that are registered to the object.

Techniques for feature extraction

Disclosed here are methods and systems that relate to determining an orientation of an object. The orientation of the object may be represented by an Euler angle which identifies a rotation of the object from a reference frame. The methods and systems may rely on readings collected from two or more barometric pressure sensors to estimate an altitude difference between the pressure sensors. The methods and systems may calculate the Euler angle based on the altitude difference.

Barometric pressure sensor based orientation measurement

Aspects of the present disclosure provide techniques for constructing a scalable model of an indoor space using crowd-sourced inertial navigation system (INS) signals from mobile devices. By tracking INS signals from a number of participating users, the user's trajectories can be estimated as they move their mobile devices indoors. The estimated trajectories can be scored against similar routes taken by other users. Routes with the highest scores are then laid out over a map of the indoor space to identify areas most often traveled to and from landmarks and distances between the landmarks.

Crowd-sourcing indoor locations

Systems and methods for detecting an attitude of a device are disclosed. The system includes a processing system including at least a sensor processor and an application processor, which are distinct. The system further includes a memory system including one or more computer-readable media. The computer-readable media contains instructions that, if executed by the processing system, cause the system to perform operations. The operations include executing an application using the application processor, and receiving raw sensor readings from one or more sensors, using the sensor processor. The operations also include determining, using the sensor processor, a processed sensor metric comprising attitude data, and transmitting the processed sensor metric to the application.

Hardware attitude detection implementation of mobile devices with MEMS motion sensors

Growing community of inventors

Palo Alto, CA, United States of America

Edward Y Chang