Abstract: The present disclosure relates to systems and methods to implement efficient high-bandwidth shared memory systems particularly suited for parallelizing and operating large scale machine learning and AI computing systems necessary to efficiently process high volume data sets and streams.

Abstract: The present disclosure relates to systems and methods to implement efficient high-bandwidth shared memory systems particularly suited for parallelizing and operating large scale machine learning and AI computing systems necessary to efficiently process high volume data sets and streams.

Abstract: Systems and methods for fully-automated contact-less biometric measurement include receiving a stream of image frames from an internet capable device. Frame data is extracted from the stream of image frames. A predictive load balancer selects a worker server based on a load prediction. An interface delivery server communicates the frame data from the internet capable device to the worker server. A set of patient face images captured by the frame data are extracted. A facial recognition machine learning model determines a patient identity associated with the set of patient face images of the frame data. A shared memory system stores the set of patient face images. The worker server determines biometric measurements based on the set of patient face images in the shared memory system using independent biometric data processing pipelines with shared access to the shared memory system for inter-process communication.

Abstract: The present invention relates to systems and methods suitable for partitioning processes between devices. In particular, the present invention relates to partitioning client devices and server devices based on the performance and available resources of the respective devices to efficiently execute artificial intelligent, machine learning, and other processes.

Abstract: Systems and methods for fully-automated contact-less biometric measurement include receiving a stream of image frames from an internet capable device. Frame data is extracted from the stream of image frames. A predictive load balancer selects a worker server based on a load prediction. An interface delivery server communicates the frame data from the internet capable device to the worker server. A set of patient face images captured by the frame data are extracted. A facial recognition machine learning model determines a patient identity associated with the set of patient face images of the frame data. A shared memory system stores the set of patient face images. The worker server determines biometric measurements based on the set of patient face images in the shared memory system using independent biometric data processing pipelines with shared access to the shared memory system for inter-process communication.

Abstract: The present disclosure relates to systems and methods to implement efficient high-bandwidth shared memory systems particularly suited for parallelizing and operating large scale machine learning and AI computing systems necessary to efficiently process high volume data sets and streams.

Abstract: A method and system can comprise a software architecture that allows different applications in the same or different communications protocols to interact with shared resources. More specifically, code for a computer program may be written to increase the amount of code that is generic to (i.e., shared by) more than one application or communications protocol and reduce the amount of code that handle application-specific or protocol-specific actions. In one embodiment, a transaction may be broken down into a set of discrete actions. The discrete actions may include functions that are common to more than one application. These functions may be part of the shared resources. Each application may have its own set of software plug-in modules for the discrete actions.

Abstract: Systems and methods for improving the performance of a data storage and retrieval system by enabling dynamic switching from one internal data structure to another in response to detecting conditions indicating that a switch would improve performance. In one embodiment, a network proxy implements a cache using a first internal data structure. The caches objects comprise Web pages, and the cache keys comprise URLs corresponding to the Web pages. The proxy monitors cache usage and periodically determines costs associated with usage of the first data structure and an alternative data structure. If the costs associated with the alternative data structure are less than the costs associated with the first data structure, the proxy crates the alternative data structure, migrates data from the first data structure to the alternative data structure, begins using the alternative data structure for the cache, and deletes the first data structure.

Abstract: A method and system can comprise a software architecture that allows different applications in the same or different communications protocols to interact with shared resources. More specifically, code for a computer program may be written to increase the amount of code that is generic to (i.e., shared by) more than one application or communications protocol and reduce the amount of code that handle application-specific or protocol-specific actions. In one embodiment, a transaction may be broken down into a set of discrete actions. The discrete actions may include functions that are common to more than one application. These functions may be part of the shared resources. Each application may have its own set of software plug-in modules for the discrete actions.

Abstract: A method and system can comprise a software architecture that allows different applications in the same or different communications protocols to interact with shared resources. More specifically, code for a computer program may be written to increase the amount of code that is generic to (i.e., shared by) more than one application or communications protocol and reduce the amount of code that handle application-specific or protocol-specific actions. In one embodiment, a transaction may be broken down into a set of discrete actions. The discrete actions may include functions that are common to more than one application. These functions may be part of the shared resources. Each application may have its own set of software plug-in modules for the discrete actions.

Abstract: Systems and methods are described which allow data to be packaged and optimized for distribution to a certain device. These systems and methods may include separating incoming data into portions, and converting each of these portions into multiple formats. Each of these portions can then be encapsulated in a packet corresponding to one portion of the data represented in one format. A packet may then be selected for delivery to a device based upon the characteristics of that device at any particular moment. In this manner, the delivery of the data may be customized according to many different criterion without impeding the delivery or usage of that data.

Abstract: Systems and methods for improving the performance of a data storage and retrieval system by enabling dynamic switching from one internal data structure to another in response to detecting conditions indicating that a switch would improve performance. In one embodiment, a network proxy is implements a cache using a first internal data structure. The cache objects comprise Web pages, and the cache keys comprise URLs corresponding to the Web pages. The proxy monitors cache usage and periodically determines costs associated with usage of the first data structure and an alternative data structure. If the costs associated with the alternative data structure are less than the costs associated with the first data structure, the proxy creates the alternative data structure, migrates data from the first data structure to the alternative data structure, begins using the alternative data structure for the cache, and deletes the first data structure.

Abstract: A method and system can comprise a software architecture that allows different applications in the same or different communications protocols to interact with shared resources. More specifically, code for a computer program may be written to increase the amount of code that is generic to (i.e., shared by) more than one application or communications protocol and reduce the amount of code that handle application-specific or protocol-specific actions. In one embodiment, a transaction may be broken down into a set of discrete actions. The discrete actions may include functions that are common to more than one application. These functions may be part of the shared resources. Each application may have its own set of software plug-in modules for the discrete actions.

Abstract: A method for operating a display having substrates and a plurality of capacitors formed at predetermined locations between the substrates includes measuring a capacitance for each of the plurality of capacitors, determining a cell gap for each of the plurality of capacitors in response to the capacitance for each of the plurality of capacitors, determining a cell gap relationship between the substrates in response to the cell gap for each of the plurality of capacitors and in response to the predetermined locations on the display, and determining a first intensity compensating value for a first pixel on an active region of the display in response to the cell gap relationship between the substrates and in response to a location of the first pixel on the display.

Abstract: The present invention provides a visual display including a high resolution miniature display compatible with VLSI technology and an optical system such as an optical magnifier used to enlarge the images display on the miniature display to be visible to the naked eye. The miniature display includes a VLSI backplane having an array of display elements monolithically formed with its driving circuit on a single crystalline semiconductor. Signal processing circuit or a microprocessor used to process image signals for the display may also be formed monolithically with the array and its driving circuit. The array may be designed using a silicon software compiler program to have randomly displaced elements or super-pixels for reducing image aliasing. The array may also be designed to have display elements positioned and scaled to compensate for the optical distortion introduced by the magnifier.

Abstract: An apparatus for redirecting physical energy includes a substrate defining a first boundary of a region, a first electrode defining a second boundary of the region, the second boundary disposed opposite to the first boundary, a second electrode adjacent to the first boundary for cooperating with the first electrode to apply a non-uniform electric field to the region, the non-uniform electric field having electrical field intensities simultaneously including a first electric field intensity and a second electric field intensity, and a layer of material disposed in the region, the layer having a variable index of refraction responsive to the electric field intensities of the non-uniform electric field, the variable index of refraction including a first index of refraction in response to the first electric field intensity and a second index of refraction in response to the second electric field intensity.

Abstract: A system having a display including display circuitry operating at a resonant frequency includes a tunable component, a resonant driver circuit coupled to the tunable component and to the display circuitry, for driving the display circuitry at an output frequency, a power consumption sensor coupled to the resonant driver circuit, for determining the power consumption of the resonant driver circuit, and a control logic coupled to the tunable component and to the power dissipation sensor, for monitoring the power consumption of the resonant driver circuit and for tuning the tunable component in response to the power consumption.

Abstract: A color microdisplay utilizes diffraction gratings to provide an array of high efficiency color pixels. The microdisplay includes a semiconductor substrate and source of light disposed adjacent thereto. A cover plate may be disposed above the substrate and has a layer of conductive material on a surface of the cover plate opposite the substrate. An optically active material, such as liquid crystal material, may be disposed between the substrate and the cover plate. An array of pixels are formed on the substrate. The pixel array includes an array of diffraction grating elements. Each element includes one or more diffraction gratings. The pitch of each diffraction grating can be a function of the angle of the incident light and the desired diffraction output spectrum. An optical system directs the diffracted light from each grating through the optically active material into viewing optics.

Abstract: The present invention provides a visual display including a high resolution miniature display compatible with VLSI technology and an optical system such as an optical magnifier used to enlarge the images display on the miniature display to be visible to the naked eye. The miniature display includes a VLSI backplane having an array of display elements monolithically formed with its driving circuit on a single crystalline semiconductor. Signal processing circuit or a microprocessor used to process image signals for the display may also be formed monolithically with the array and its driving circuit. The array may be designed using a silicon software compiler program to have randomly displaced elements or super-pixels for reducing image aliasing. The array may also be designed to have display elements positioned and scaled to compensate for the optical distortion introduced by the magnifier.

Abstract: A color microdisplay utilizes diffraction gratings to provide an array of high efficiency color pixels. The microdisplay includes a semiconductor substrate and source of light disposed adjacent thereto. A cover plate may be disposed above the substrate and has a layer of conductive material on a surface of the cover plate opposite the substrate. An optically active material, such as liquid crystal material, may be disposed between the substrate and the cover plate. An array of pixels are formed on the substrate. The pixel array includes an array of diffraction grating elements. Each element includes one or more diffraction gratings. The pitch of each diffraction grating can be a function of the angle of the incident light and the desired diffraction output spectrum. An optical system directs the diffracted light from each grating through the optically active material into viewing optics.