127 results about "General purpose computer" patented technology

A method, apparatus, non-transitory computer readable medium, apparatus, and system for image generation include obtaining a sketch input depicting an object, processing the sketch input to obtain sketch guidance, and generating a synthesized image based on the sketch guidance using an image generation model, where the synthesized image depicts the object from the sketch input.

A method, apparatus, non-transitory computer readable medium, and system for image generation include obtaining, via a user interface, a reference image and generating, using an image generation model, a synthesized image based on the reference image. A layered image is generated including the synthesized image in a first layer of the layered image and the reference image in a second layer of the layered image. The layered image is then presented for display in the user interface.

The present disclosure relates to systems, methods, and non-transitory computer-readable media that recolors a digital design according to colors of a digital image and further generates an enhanced recolored digital design. In particular, in some embodiments, the disclosed systems identify a digital image for recoloring a digital design and recolors the digital design utilizing a color affine transformation algorithm to generate a recolored digital design. Further, in some embodiments, the disclosed systems generate the enhanced recolored digital design by transforming one or more colors of the recolored digital design to be within a range of the colors of the digital image utilizing a convex hull projection method. Moreover, in some embodiments, the disclosed systems further enhance the recolored digital design utilizing a contrast enhancement algorithm to modify luminescence values.

Described herein are machine learning based system, method, and non-transitory computer readable medium for determining three-dimensional (3D) information of a structure of a patterned substrate. The 3D information can be determined using a neural network to convert a two-dimensional image to a 3D image. In a method, the neural network is trained by supplying simulated images of a sample with varied parameters to model a real fabricated IC structure. The trained network generates a depth map from newly supplied images of a sample to predict defect location The depth map may be converted to a binary map to predict defect dimension and location that may be used to guide inspection of a sample using an inspection tool.

The present disclosure is directed toward systems, methods, and non-transitory computer-readable media for generating context engine outputs by utilizing an interpreter purpose-built to execute code generated by large language models. For example, the disclosed systems generate computer code executable for responding to a query by utilizing a large language model. In addition, the disclosed systems execute the model-generated computer code utilizing an interpreter integrated with the context engine that further includes swappable logic interchangeable across multiple executors. Moreover, the disclosed systems can further generate as part of executing the computer code utilizing the interpreter, a first context engine output by implementing the interpreter at a first executor. Additionally, the disclosed systems can further generate a second context engine output by implementing the interpreter at a second executor.

A method, communication device and non-transitory computer storage media configured for reporting sensing, where sensing data corresponding to a sensing service is reported based on information on a sensing policy. The sensing data indicates features of a sensed object. Reporting involves a control plane network function, which includes access and mobility management function (AMF), session management function (SMF), and policy control function (PCF). The sensing policy may define various aspects such as service type, object type, data collection manner and frequency, and reporting parameters. Sensing data can be raw or processed, and services may be triggered by applications or network functions. Information on the sensing policy is received during network registration or after service triggering.

An action detection system (10) includes an action specification unit (18) configured to start specification of at least one action ID based on at least part of skeleton information extracted from video data that is a captured video of a user in response to detection of the user visiting checkout machine determined in advance, a determination unit (20) configured to determine whether or not an action sequence including the specified at least one action ID corresponds to registration action sequences registered in advance, and a processing control unit (21) configured to execute processing determined in advance in accordance with a determination result.

To provide a mechanism capable of easily inserting an AI image generated by a desired keyword into a desired area of an image.SOLUTION: A CPU 170 of a general-purpose computer 102 designates an area into which an AI image is inserted in a document on the basis of a vector 305 when the vector is inputted in a document image being displayed by a user. After that, the CPU 170 displays a window, an AI image generation request including a keyword 306 is transmitted to an image generation server 101 when the keyword 306 of the AI image is inputted to the window by a user, acquires an AI image generated by inputting the keyword 306 to a learned model 164 from the image generation server 101, and inserts the acquired AI image into the designated area in the document.SELECTED DRAWING: Figure 2

Disclosed are systems, methods, and non-transitory computer-readable storage media for setting a reminder triggered by a target device. A requesting device sends a request to a server to set a notification triggered by a target device. The request includes parameters, such as a location and a condition that define when the notification is triggered. The server sends instruction to the target device to set the notification based on the parameters. When the condition such as arrival is met by the target device in relation to the location the target device sends a message to the server that the notification has been triggered. The target device can set a geo-fence to determine the position of the target device in relation to the location, and the requesting user can dictate the size of the geo-fence. The server sends a message to the requesting device that the notification has been triggered.

Disclosed in the present application are an image acquisition method, an imaging apparatus, an electronic device, and a non-transitory computer-readable storage medium. The image acquisition method includes: controlling exposure of a pixel array; and performing an interpolation process on a first colored original image and a second colored original image according to a first panchromatic original image, and fusing an interpolated image and the first panchromatic original image to obtain a target image with same resolution as resolution of the pixel array.

Disclosed are methods, systems and non-transitory computer readable memory for trust-aware error detection, correction, and explainability. For instance, a method may include processing a test instance through a machine learning model to obtain a set of inferences; detecting errors in the set of inferences; automatically correcting the errors in the set of inferences and / or updating the machine learning model; determining post-hoc explanations into a cause of the errors; and outputting the post-hoc explanations to a user.

Systems, methods and non-transitory computer readable media for artificially generating translated media streams are provided. A media stream of an individual speaking in an origin language may be received. The individual may be associated with a particular voice. The received media stream may be analyzed to determine characteristics of the particular voice of the individual. A synthesized voice may be determined based on the determined characteristics of the particular voice of the individual. The synthesized voice may sound substantially identical to the particular voice. A translated media stream that includes, for each word spoken in the origin language in the media stream, a respective at least one word in the target language articulated using the synthesized voice, may be generated.

The invention belongs to the field of general computers, and provides a hydraulic data processing method and system based on artificial intelligence, and the method comprises the steps: constructing a first time sequence and a second time sequence corresponding to nodes; calculating a time difference vector by using a variational encoder; constructing a graph structure in combination with the physical connection relationship between the monitoring points, and determining the structural representation of each node; performing image segmentation on the remote sensing image data, extracting image region features corresponding to node positions, splicing the image region features with meteorological data at the same position, and determining image-meteorological representation of each node; for each node, splicing the structure representation output by the graph convolutional neural network with the image-weather representation to form fusion representation; a disturbance vector of a preset amplitude is injected into the fusion representation, disturbance representation is obtained, and the disturbance direction is the gradient direction of the loss function relative to the fusion representation; and jointly inputting the fusion representation and the disturbance representation into a dichotomy neural network, and outputting a judgment result about whether abnormal mutation occurs or not.

A parking assistance apparatus includes: an image acquirer that acquires a camera image; a feature point detector that extracts feature points from the camera image; a feature point selector that selects a feature point to be registered on a map by evaluating the feature points in learning travel and a parking route and a parking position of the vehicle are registered on the map; and a vehicle controller that parks the vehicle based on the map, in which the feature point selector varies a priority for registration of the feature points or the number of feature points for registration in accordance with a position on the parking route, or selects, at the position on the parking route, the feature point to be registered on the map, based on a position of the camera or a relative position of the feature point with respect to an optical axis direction.

This disclosure describes a system, method, and non-transitory computer-readable medium for image search-based object detection of utility assets in image databases. The method includes receiving an input image of a utility asset and a query bounding box representing an image-based object query. Bounding boxes of objects represented in the input image are generated based on the input image and the query bounding box, in which anchoring boxes corresponding to object classifications are identified from the bounding boxes. A textual label is determined for a selected subset of anchoring boxes. An image embedding representing the region is encoded, and image tokens are generated based on the encoded image embedding. Output images of other utility assets relevant to the image-based object query are identified from images in an image database, based on at least one of (i) the encoded image embedding, (ii) the image tokens, or (iii) the textual label. The output images are provided for output.

A system comprising one or more processors and one or more non-transitory computer-readable media storing computing instructions that, when executed on the one or more processors, cause the one or more processors to perform operations comprising: determining, using a predictive analysis, an estimated collection time to collect one or more items of an order; receiving a mobile check-in from an electronic device indicating that a pickup is en-route to pick-up the order from the store; tracking a location of the pickup; periodically determining an estimated travel time for the pickup to travel to the store from the location, as tracked; and transmitting instructions for the one or more employees of the store to begin collecting the one or more items of the order on a store interface Other embodiments are disclosed herein.

This disclosure describes one or more implementations of systems, non-transitory computer-readable media, and methods that perform language guided digital image editing utilizing a cycle-augmentation generative-adversarial neural network (CAGAN) that is augmented using a cross-modal cyclic mechanism. For example, the disclosed systems generate an editing description network that generates language embeddings which represent image transformations applied between a digital image and a modified digital image. The disclosed systems can further train a GAN to generate modified images by providing an input image and natural language embeddings generated by the editing description network (representing various modifications to the digital image from a ground truth modified image). In some instances, the disclosed systems also utilize an image request attention approach with the GAN to generate images that include adaptive edits in different spatial locations of the image.

Systems, methods, and non-transitory computer-readable media for organizing graphical user interface (GUI) elements of a GUI of a computing device. An electronic processor generating a plurality of team GUI elements corresponding to a plurality of team tables. Transmitting the plurality of team GUI elements for display on the computing device via the GUI. Receiving a participant selection of a team GUI element of the plurality of team GUI elements that are displayed, wherein the team GUI element includes one or more records. Populating a first window of the GUI based on the participant selection of the team GUI element. Receiving a participant selection of a record of the one or more records of the participant selected team GUI element. Automatically adding the record to a roster GUI element of a second window of the GUI based on an attribute of a field of the record.

Methods, systems, and non-transitory computer readable storage media are disclosed for managing computing systems to classify and modify digital content items to satisfy digital data requirements of data policies. For example, the content classification system validates, enforces, and remediates digital data content corresponding to digital data requirements of a data policy based on data types covered by the data policy. The disclosed systems generate classifications for digital content items by accessing digital content items and generating mappings between the digital content items and a data policy. The disclosed systems utilize the mappings and digital data requirements of the data policy to determine whether the digital content items violate one or more elements of the data policy. The disclosed systems can perform various downstream operations to remediate the data policy violations, such as by causing various computing devices to modify the violating digital content items.

The present disclosure relates to systems, methods, and non-transitory computer-readable media that generates object masks for digital objects portrayed in digital images utilizing a detection-masking neural network pipeline. In particular, in one or more embodiments, the disclosed systems utilize detection heads of a neural network to detect digital objects portrayed within a digital image. In some cases, each detection head is associated with one or more digital object classes that are not associated with the other detection heads. Further, in some cases, the detection heads implement multi-scale synchronized batch normalization to normalize feature maps across various feature levels. The disclosed systems further utilize a masking head of the neural network to generate one or more object masks for the detected digital objects. In some cases, the disclosed systems utilize post-processing techniques to filter out low-quality masks.

In some aspects, a non-transitory computer readable medium is disclosed that has an application stored thereon that when executed by a processor of an electronic device causes the processor to perform a process. The process can include receiving motion data from a motion sensor, the motion data being indicative of motion by a user while wearing the motion sensor. The process can also include transmitting the motion data to a server, receiving from the server a time that the motion data indicates a foot of the user experienced a loading force which risked adversely impacting an existing foot ulcer or causing a new foot ulcer, determining a geographic location of the motion sensor at the time that the foot experienced the loading force, displaying the time and the geographic location, and requesting that the user identify an activity engaged in by the user at the time and the geographic location.

The invention discloses a communication protocol signal acceleration processing method based on a general-purpose computer, and the method achieves the quick signal detection, signal processing, recognition and positioning of IQ intermediate frequency data of a communication protocol signal through splitting the processing link of the communication protocol signal based on a distributed computing technology and an efficient communication protocol processing algorithm. Therefore, the timeliness of communication protocol signal processing is obviously improved.

A robot control system includes a robot controller, a data transmission module, a servo drive module, a safety unit, a demonstrator, and a power module. The servo drive module is connected to the robot controller via the data transmission module and receives a movement instruction to drive a robot to move. The safety unit is connected to the robot controller and the servo drive module via the data transmission module and turns off, upon receiving an abnormal input signal or a failure signal, the servo drive module and transmits the abnormal input signal or the failure signal to the robot controller. The demonstrator receives a terminal video signal and a first control interaction signal from the robot controller, and sends a second control interaction signal to the robot controller to control operation of the robot controller. The power module is electrically connected to the robot controller and the safety unit.

Systems and methods for mathematical modeling of the hardware and software stack of commodity computer platforms are provided, enabling provable guarantees on memory, device, and program execution. This approach addresses the technical problem of reliance on system agents that rely on implicit trust in the operating environment, which can be exploited by sophisticated attackers using complex threats such as memory access exploits and code / data integrity exploits. The solution provides a proactive, mathematically-backed security solution that eliminates entire classes of cyberattacks by design, ensuring realizable guarantees on commodity computer platforms running hardware and software stack elements at the lowest operating level. This approach has significant advantages over current reactive cybersecurity methods, including reduced complexity and overhead, and increased confidence in the integrity of the system. The solution's main uses include providing mathematically-backed security and availability guarantees for critical infrastructure, financial institutions, and other organizations vulnerable to cyberattacks.

The present disclosure relates to configuring computing nodes in a distributed ledger environment (e.g., a blockchain environment). Smart contracts and blockchains are used in a wide variety of application fields. Conventionally, smart contracts are executed on a general-purpose computer, which often underperform for a variety of tasks associated with implementing a blockchain network. The present disclosure describes a novel smart contract system implementable on nodes of a blockchain network in order to provide improved or optimized performance. For example, the smart contract system can leverage instruction-level parallelism as well as transaction-level parallelism during smart contract processing in order to boost its execution performance. Further, the smart contract system can be configured and adapted to different workloads in order to remove bottlenecks. The processes described herein can additionally be implemented on unique hardware uniquely suited for the parallelism involved in processing blockchain transactions.

A system-on-chip-in-the-loop (SoCIL) for closed-loop simulation of a control system includes a system-on-chip (SoC) field-programmable gate array (FPGA) board. The SoC FPGA board includes a target controller that is an electronic control unit (ECU). The target controller includes one or more target controller soft processors. The SoC FPGA includes a real-time computer that is a general-purpose computer. The real-time computer includes one or more real-time soft processors. The SoC FPGA includes one or more high-frequency simulation modules including a clock speed having a frequency in the Megahertz range and one or more high-frequency physical plant models. The one or more target controller soft processors of the target controller, the one or more real-time soft processors of the real-time computer, and the one or more high-frequency physical plant models of the one or more high-frequency simulation modules are in electronic communication with one another by an on-chip communication bus protocol.

Systems, methods, and non-transitory computer-readable media for generating insights for increasing revenue including receiving past and current data on various metrics for a plurality of sale representatives for an organization; training a model to predict a potential revenue attainment based on the received data; calculating the potential revenue attainment for each of the plurality of sale representatives; selecting one representative; determining an impact of each performance metric, activity metric, competency metric, and execution / engagement metric on the potential revenue attainment; identifying one performance metric having the most impact on the potential revenue attainment; determining a correlation coefficient between each root cause metric and the identified one performance metric; identifying one root cause metric having the most impact on the identified one performance metric; and providing one or more recommendations for the one representative on an interactive sales dashboard based on the identified one root cause metric.

The described systems and related test methods provide a significant improvement in test accuracy, as well as a reduction in test time for system-level electromagnetic environmental effects (E3) testing. The systems may include a real-time spectrum analyzer, a network analyzer, and a switching / filtering / coupling network, controlled by an information processing device such as a general-purpose computer. These systems can scan all frequencies at each receiver.

A method, a sensor system, and a non-transitory computer readable medium are provided. Embodiments comprise receiving a plurality of incident signals reflected from a target at a plurality of receiver antennas. A processor identifies a subset of receiver antennas from the plurality of receiver antennas and performs angle of arrival (AOA) processing on respective incident signals of the subset of receiver antennas. The processor determines an angle of arrival of the target based on the AOA processing.