International Business Machines Corporation

Steven Edward Millman

Steven Lorenz Wright

Kai R Schleupen

Kenneth C Ho

Richard Ian Kaufman

Paul F Greier

Alphonso Philip Lanzetta

Nirmit V Desai

Heather D Achilles

Kevin Wilson Warren

Shahrokh Daijavad

Wendy Chong

Michael Patrick Mastro

Carmelo I Uria

Steven T Berman

Ron Ridgeway

Chai Wah Wu

Gerhard Robert Thompson

Kiyoshi Maruyama

Catherine Helen Crawford

Satish Gupta

Carl Michael Jefferson

Roger F Hoyt

Halil Burhan Bakoglu

Thomas Kimber Worthington

Chung-wai Yue

Marc Herbert Brodsky

Christopher P Huff

Peter F Mycue

Kelvin Kakugawa

Adam E Grant

Randy M Fussell

A notification message is received by a communication device The notification message includes a message identifier and a digital signature generated by an original sender of the notification message. That an integrity of message content in the notification message and an original sender of the notification message is verified is determined using the digital signature. Responsive to determining that the notification message is verified, the notification message is forwarded by the communication device using a peer-to-peer communication channel over a mesh network.

Delivering messages to offline devices using peer-to-peer communication

A trained base model is distributed to a set of nodes. From a first node in the set of nodes, a first set of meta-metrics resulting from a transfer learning operation on the trained base model at the first node is collected. The transfer learning at the first node uses first local data available at the first node. The first node is clustered in a cluster with a second node from the set of nodes, in response to a meta-metric in the first set of meta-metrics being within a tolerance value of a corresponding meta-metric in a second set of meta-metrics collected from the second node. A normalized set of model parameters is constructed after an iteration of transfer learning or local learning at the first and second nodes. The normalized set of model parameters is distributed to the first node and the second node in the cluster.

Transfer learning without local data export in multi-node machine learning

A video processing method and system for generating a foveated video display with sections having different resolutions uses a network channel for communicating video images having video sections of different resolutions, and includes a video transmission system for processing and transmitting the received video images over the network channel. The system assigns a larger portion of the network channel's bandwidth to a video section with higher resolution. Further, the system includes a video receiving system for receiving and seamlessly combining the first and second video sections of different resolutions to form an output video image on a display device, and a control unit for sending one or more video control parameters to the video transmission system to control capturing, transmitting and processing of the video images.

System and method for selective image capture, transmission and reconstruction

When run on high pixel density monitors, software applications, which are written to be legible at all resolutions benefit greatly from the high pixel density. Other applications, especially those containing resources whose dimensions and placement are described in terms of numbers of pixels (e.g., a bitmap, text), may have reduced legibility. A method is described which facilitates legibility of both classes of software applications when run on high pixel density monitors.

Method for improving the viewing of small icons and small text on high resolution displays

Viewing angle characteristics of a liquid crystal display (LCD) are improved by reducing the number of subpixels in an image with mid-tone luminance values. In a preferred embodiment, a first table of entries associating subpixel intensity values and subpixel luminance values for a LCD in at least one viewing angle direction is provided. A target intensity value is determined from the first table, corresponding to the average subpixel luminance over a small number of adjacent subpixels. A second table of entries associates the target intensity values with intensity values above and below the target. The adjacent subpixel intensity values are modified according to the second table, thereby reducing the number of subpixels with mid-tone luminance values. The subpixel data is preferably processed within a portion of an application-specific integrated circuit (ASIC), contained within the display module.

Method and apparatus for adjusting subpixel intensity values based upon luminance characteristics of the subpixels for improved viewing angle characteristics of liquid crystal displays

The present invention provides for a software routine executable by an image display system that controls at least one video timing signal (for example, the pixel clock signal) supplied to a display subsystem. More specifically, the software routine, adjusts the video timing signal in response to detection that the system is switching between power modes. The software routine preferably updates the video timing signal such that the system conserves power in response to detection that the system is switching from AC-powered operation to limited DC-powered operation.

Method and apparatus for adjusting video refresh rate in response to power mode changes in order to conserve power

A method is provided for rendering display blocks on a video display device of a computer processing system having an existing display driver. The method includes the step of assigning each of a plurality of n video adapters to an m,display subregion of the display device. A call to a draw function of the existing display driver is received. The draw function corresponds to a particular shape to be drawn and includes starting and ending coordinates. The m,display subregion that drawing of the shape is to begin on is calculated from the starting coordinates. A function of the existing display driver is called by calling an address stored in the filter driver of a routine corresponding to the called function. The called function corresponds to the received draw function. A segment of the shape is rendered on the calculated m,display subregion by the n,display adapter corresponding thereto. A traversal is made to the n,video adapter corresponding to a next segment of the shape. The next segment of the shape is rendered on the m,display subregion corresponding to the traversed n,video adapter. The traversal and second rendering step are repeated until all segments of the shape have been rendered.

Method for rendering display blocks on display device

Allows a video broadcast viewer to pause at anytime while viewing a program, and upon returning to be able to view the intervening segment. The video received during the pause period is stored and available for recall upon user command. The storage medium is circular in so much that the memory, upon becoming filled to capacity continues to write incoming information data over the previously stored data. The storage circuits employs a sequential access storage device and/or a direct access storage device. The storage circuit has a high speed memory input buffer and a high speed memory output buffer to account for the relatively long READ and WRITE access times of the storage device employed. At the end of the pause period, the user can view the stored video. The user has a choice between catching up to the regular broadcast or watching the remaining program video in a delay mode. The stored video may be viewed intermittently or continuously in regular, slow or fast motion, under user direct or remote VCR-type of video function control.

Interruption tolerant video program viewing

A self-contained CD-ROM player which can be installed and used as simply as a VCR. The CD-ROM player includes a CD-ROM drive and supporting hardware and software to activate and execute a program on a CD-ROM disk. It is not necessary to explicitly issue commands to activate and use the disk. The user interface is completely intuitive and simple even for someone who is completely computer illiterate. This player allows everyone, both proficient computer users and those not so proficient, to use the vast library of CD-ROM programs that have been written for personal computers.

User friendly compact disk (CD) read only memory (ROM) player

A self-diagnostic method and apparatus is described for indicating the likelihood of an impending crash of a magnetic transducer or head with a rotating magnetic medium by dynamically and periodically calculating overwrite efficiency for performing a desired control operation when it exceeds a preselected value denoting dangerous proximity of the transducer to the surface of the medium.

Magnetic transducer crash anticipation and response method and apparatus

Growing community of inventors

Spring Valley, NY, United States of America

Steven Edward Millman