Abstract: Mechanisms are provided to implement a machine learning training model. The machine learning training model trains an image generator of a generative adversarial network (GAN) to generate medical images approximating actual medical images. The machine learning training model augments a set of training medical images to include one or more generated medical images generated by the image generator of the GAN. The machine learning training model trains a machine learning model based on the augmented set of training medical images to identify anomalies in medical images. The trained machine learning model is applied to new medical image inputs to classify the medical images as having an anomaly or not.

Abstract: Embodiments described herein relate to methods, devices, and computer-readable media to determine a compression setting. An input image may be obtained where the input image is associated with a user account. One or more features of the input image may be determined using a feature-detection machine-learning model. A compression setting for the input image may be determined using a user-specific machine-learning model personalized to the user account based on the one or more features in the input image. The input image may be compressed based on the compression setting.

Abstract: Helper neural network can play a role in augmenting authentication services that are based on neural network architectures. For example, helper networks are configured to operate as a gateway on identification information used to identify users, enroll users, and/or construct authentication models (e.g., embedding and/or prediction networks). Assuming, that both good and bad identification information samples are taken as part of identification information capture, the helper networks operate to filter out bad identification information prior to training, which prevents, for example, identification information that is valid but poorly captured from impacting identification, training, and/or prediction using various neural networks. Additionally, helper networks can also identify and prevent presentation attacks or submission of spoofed identification information as part of processing and/or validation.

Abstract: The present disclosure relates to systems, methods, and non-transitory computer readable media for accurately and flexibly generating modified digital images utilizing a novel swapping autoencoder that incorporates scene layout. In particular, the disclosed systems can receive a scene layout map that indicates or defines locations for displaying specific digital content within a digital image. In addition, the disclosed systems can utilize the scene layout map to guide combining portions of digital image latent code to generate a modified digital image with a particular textural appearance and a particular geometric structure defined by the scene layout map. Additionally, the disclosed systems can utilize a scene layout map that defines a portion of a digital image to modify by, for instance, adding new digital content to the digital image, and can generate a modified digital image depicting the new digital content.

Abstract: A method including receiving a first set of one or more street level images of houses into a Machine Learned Model trained with a second set of street level images with one or more exterior features of the houses labeled, identifying the one or more exterior features in the first set of street level images by way of the Machine Learned Model, and quantifying and outputting the counts and/or two-dimensional areas for each of the identified exterior features in the first set of one or more street level images is described. Non-transitory, computer-readable storage media having instructions for executing the method steps by one or more processors as well as computer or computer systems capable of performing the method steps are also described.

Abstract: A method includes storing at least one image performance score for each of a set of images, the set of images comprising a plurality of subsets of images, each subset corresponding with a different web page of a plurality of web pages, the at least one image performance score for an image indicating a likelihood that a user will interact with the image; determining a web page score for each of the plurality of web pages based on one or more image performance scores of the subset of images that corresponds with the web page; receiving a query comprising one or more keywords or images; selecting a set of web pages by applying a search engine machine learning model to the one or more keywords and the web page score for each of the plurality of web pages; and presenting the set of web pages at a computing device.

Abstract: A model generating apparatus includes a measurement unit and a change unit. The measurement unit measures a size of an object appearing on an image included in data set. The change unit changes, based on a distribution of the size, a layer to be connected to a detection unit of a Convolutional Neural Network (CNN) for detecting an object appearing on the image, among layers for extracting feature included in the CNN.

Abstract: An image synthesis device according to a disclosed embodiment has one or more processors and a memory which stores one or more programs executed by the one or more processors. The image synthesis device includes a first artificial neural network provided to learn each of a first task of using a damaged image as an input to output a restored image and a second task of using an original image as an input to output a reconstructed image, and a second artificial neural network connected to an output layer of the first artificial neural network, and trained to use the reconstructed image output from the first artificial neural network as an input and improve the image quality of the reconstructed image.

Abstract: Systems, devices, methods and instructions are described for detecting GAN generated images. On embodiment involves receiving an images, generating co-occurrence matrices on color channels of the image, generating analysis of the image by using a convolutional neural network trained to analyze image features of the images based on the generated co-occurrence matrices and determining whether the image is a GAN generated image based on the analysis.

Abstract: Methods and systems for image filtering include detecting a distortion level of input images, using a distortion detection model that is trained using confidence values generated by a pre-trained image classifier with a set of distorted training images. An analysis is performed on input images having a detected distortion level that is lower than a threshold, with input images having an above-threshold detected distortion level being filtered out.

Abstract: Systems and methods for producing high-quality image segmentation masks by efficiently augmenting low-quality segmentation masks initially obtained by lightly trained neural networks are disclosed. Sequences of low-quality segmentation masks are transformed into high-quality segmentation masks by first identifying groups of pixels that are consistently segmented in at least three successive video frames. The pixels surrounding the consistently segmented pixels are then examined and assigned gradient alpha values, i.e., degrees of transparency. In one embodiment the alpha values are made proportional to the RGB distance of the pixel from the mean RGB value of the consistently segmented group of pixels.

Abstract: The learning device includes a metric space learning unit and a case example storage unit. The metric space learning unit learns a metric space including feature vectors extracted from attributed image data, for each combination of different attributes, using the attributed image data to which attribute information is assigned. The case example storage unit computes the feature vector from the image data for case example to store the computed feature vector as a case example associated with the metric space, and stores additional information associated with the case example.

Abstract: An inspection apparatus according to one or more embodiments extracts an attention area from a target image using a first estimation model, performs a computational process with a second estimation model using the extracted attention area, and determines whether a target product has a defect based on a computational result from the second estimation model. The first estimation model is generated based on multiple first training images of defect-free products in a target environment. The second estimation model is generated based on multiple second training images of defects. The computational process with the second estimation model includes generating multiple feature maps with different dimensions by projecting the target image into different spaces with lower dimensions. The extracted attention area is integrated into at least one of the multiple feature maps in the computational process with the second estimation model.

Abstract: Methods, systems, and computer programs are presented for the generation of content in advance to enable quickly customized communications for multiple types of customers. One method includes an operation for identifying components of an image design that specifies how the components are combined to generate an image. For one or more of the identified components, variations of the components are generated using one of several generative artificial intelligence (GAI) models. The method further includes detecting a request, comprising user attributes, for the image. For one or more of the identified components, a respective variation is selected based on the user attributes, and a response image is created utilizing the image design and the one or more selected variations. Further, the response image is presented on a computer user interface.

Abstract: A method for training a super-resolution network may include obtaining a low resolution image; generating, using a first machine learning model, a first high resolution image based on the low resolution image; generating, using a second machine learning model, a second high resolution image based on the first high resolution image and an unpaired dataset of high resolution images; obtaining a training data set using the low resolution image and the second high resolution image; and training the super-resolution network using the training data set.

Abstract: The present invention discloses a high-resolution hyperspectral computational imaging method and system and a medium. The method of the present invention comprises: conducting spectral upsampling on an input RGB image Y to obtain an initial hyperspectral image X0; and inputting the initial hyperspectral image X0 into a pre-trained deep convolutional neural network guided by an imaging model, and conducting iteration computation to obtain a hyperspectral image X. The present invention can effectively achieve reconstruction of the RGB image to the high-resolution hyperspectral image and has the advantages of high reconstruction precision, high computational efficiency, little memory consumption and strong generalization ability.

Abstract: The present disclosure relates to systems, methods, and non-transitory computer readable media for training a generative inpainting neural network to accurately generate inpainted digital images via object-aware training and/or masked regularization. For example, the disclosed systems utilize an object-aware training technique to learn parameters for a generative inpainting neural network based on masking individual object instances depicted within sample digital images of a training dataset. In some embodiments, the disclosed systems also (or alternatively) utilize a masked regularization technique as part of training to prevent overfitting by penalizing a discriminator neural network utilizing a regularization term that is based on an object mask. In certain cases, the disclosed systems further generate an inpainted digital image utilizing a trained generative inpainting model with parameters learned via the object-aware training and/or the masked regularization.

Abstract: A method and apparatus for performing storage and retrieval in an information storage system cache is disclosed that uses the hashing technique with the open-addressing method for collision resolution. Insertion, retrieval, and deletion operations are limited to a predetermined number of probes, after which it may be assumed that the table does not contain the desired data. Moreover, when using linear probing, the technique facilitates maximum concurrent, multi-thread access to the table, thereby improving system throughput, since only a relatively small section is locked and made unavailable while a thread modifies that section, allowing other threads complete access to the remainder of the table.

Abstract: An example method for estimating a number of individuals present in a crowd includes generating a density map based on an image of a crowd, using a convolutional neural network, augmenting the density map with cellular signal processing data to produce an augmented density map, and estimating a number of individuals present in the crowd, based on the augmented density map.

Abstract: Embodiments of the present disclosure relate to a method, a device, and a computer program product for processing data. The method includes determining, based on acquired text, character features for a group of characters in the text and text features for the text. The method further includes determining initial visual features for the text based on the text features. The method further includes determining target visual features for the text based on the initial visual features, the character features, and the text features. The method further includes generating a target image corresponding to the text based on the target visual features. Through the method, the accuracy of conversion between text and an image is improved, the data processing efficiency is improved, and the data compression efficiency is further improved.

Abstract: A model building apparatus includes: a building unit that builds a generation model that outputs an adversarial example, which causes misclassification by a learned model, when a source sample is entered into the generation model; and a calculating unit that calculates a first evaluation value and a second evaluation value, wherein the first evaluation value is smaller as a difference is smaller between an actual visual feature of the adversarial example outputted from the generation model and a target visual feature of the adversarial example that are set to be different from a visual feature of the source sample, and the second evaluation value is smaller as there is a higher possibility that the learned model misclassifies the adversarial example outputted from the generation model. The building unit builds the generation model by updating the generation model such that an index value based on the first and second evaluation values is smaller.

Abstract: An anomaly detection device based on a generative adversarial network architecture, which uses the single-type training data composed of multiple normal signals to train an anomaly detection model. The anomaly detection device includes an encoder, a generator, a discriminator, and a random vector generator. In the training phase of anomaly detection model, the random latent vectors generated by the random vector generator are sequentially input to a generator to generate the synthesized signals with the same dimension as the normal signals. The synthesized signals are sequentially input into a discriminator to output the corresponding discriminant values. When the corresponding discriminant values are under the predetermined threshold, the corresponding synthesized signals are selected as the anomalous class training data, and the real normal signals are selected as the normal class training data.

Abstract: Provided is an embedded semantic division network including a communication module configured to receive an image captured by a camera, a memory configured to store a semantic division network (MMANet)-based program for extracting a context of the captured image, and a processor extracts the context of the captured image by selecting a convolutional neural network (CNN) processing module or a depth-wise separable convolution (DSC) processing module according to a size of a activation map in each layer of the semantic division network that includes an encoder unit and a decoder unit including at least one of the CNN processing module and the DSC processing module that are connected from an upper layer to a lower layer and reduce features of an input image.

Abstract: Provided are a training method and apparatus for an image processing image processing model, and an image processing method and apparatus. The training method comprises: acquiring a sample image and a first reference image, wherein the information quantity and resolution of the sample image are respectively lower than those of the first reference image; inputting the sample image into a generative network in an image processing model, and carrying out super-resolution processing and down-sampling processing on the sample image by means of the generative network, so as to generate and output at least one result image; determining the total image loss of the at least one result image according to the first reference image; and adjusting parameters of the generative network according to the total image loss, so that the total image loss of at least one result image output by the adjusted generative network meets an image loss condition.

Abstract: A training method for multi-output land cover classification model and a classification method are provided. The training method includes: obtaining a training data; inputting the training data into an initial model based on deep belief nets for training to obtain a multi-output land cover classification model, wherein the initial model includes N level outputs, and the N level outputs include an output set at last network layer and (N?1) level output set at any (N?1) network layers from a first network layer to a penultimate network layer of the initial model; determining a total loss according to losses of the N level outputs; performing a backpropagation based on the total loss to adjust a parameter of the initial model, N being an integer greater than or equal to 2. The gradient is not easy to disappear during backpropagation of the model, which is beneficial to improve classification accuracy.

Abstract: Generative AI systems and methods are developed to provide recommendations regarding the control, optimization, and troubleshooting of industrial equipment and manufacturing systems as determined from a range of available data sources. A consistent, semantic metadata structure is described as well as a hypothesis generating and testing system capable of generating predictive analytics models in a non-supervised or partially supervised mode. Users and/or AI agents (i.e., a form of “agentic AI”) may then subscribe to such information for the use in connection with their own manufacturing systems.

Abstract: Systems and methods for improving training processes for image and text applications are described. A first set of embeddings may be generated based on a text input, and a second set of embeddings may be generated via a convolutional neural network (CNN), based on an input image. The first set of embeddings and the second set of embeddings may be utilized to generate a third set of embeddings including one or more placeholder values to be replaced. The placeholder values may be replaced based on predicted values, to reconstruct the input text and image.

Abstract: A method includes storing at least one image performance score for each of a set of images, the set of images comprising a plurality of subsets of images, each subset corresponding with a different web page of a plurality of web pages, the at least one image performance score for an image indicating a likelihood that a user will interact with the image; determining a web page score for each of the plurality of web pages based on one or more image performance scores of the subset of images that corresponds with the web page; receiving a query comprising one or more keywords or images; selecting a set of web pages by applying a search engine machine learning model to the one or more keywords and the web page score for each of the plurality of web pages; and presenting the set of web pages at a computing device.

Abstract: Systems/techniques that facilitate machine learning generation of low-noise and high structural conspicuity images are provided. In various embodiments, a system can access an image and can apply at least one of image denoising or image resolution enhancement to the image, thereby yielding a first intermediary image. In various instances, the system can generate, via execution of a plurality of machine learning models, a plurality of second intermediary images based on the first intermediary image, wherein a given machine learning model in the plurality of machine learning models receives as input the first intermediary image, wherein the given machine learning model produces as output a given second intermediary image in the plurality of second intermediary images, and wherein the given second intermediary image represents a kernel-transformed version of the first intermediary image. In various cases, the system can generate a blended image based on the plurality of second intermediary images.

Abstract: According to one embodiment, a method is provided for video repainting performed by at least one processor of a computer device. The method includes receiving a video sequence having one or more image frames; detecting presences of a target object within the one or more image frames and determining pose condition and style shift of the detected objects; generating content representing a replacement object for the one or more image frames by applying the corresponding pose condition and style shift to the replacement object; and repainting the detected target object in the one or more image frames with the generated content.

Abstract: Embodiments relate to a deep learning/AI-based product surface quality inspection system which is accurate and reliable in product quality inspection which is a core task in an injection process among various manufacturing fields. The system can provide, to a user, better performance than non-defective/defective manufactured product classification methodology which is an existing commonly used method through a method and a system considering characteristics of a factory environment and an actual product production process for all pipelines of product quality inspection by using only a non-defective manufactured product image unlike most injection process surface inspection AIs developed to date.

Abstract: A method for processing microscope images in order to generate an image processing result comprises: implementing a convolutional neural network, wherein a first convolutional layer calculates an output tensor from an input tensor formed from a microscope image. The output tensor is input into one or more further layers of the convolutional neural network in order to calculate the image processing result. The first convolutional layer comprises a plurality of filter kernels. At least several of the filter kernels are respectively representable by at least one filter matrix with learning parameters and dependent filter matrices with implicit parameters, which are determined by means of the learning parameters and one or more weights to be learned, wherein the filter matrices with learning parameters of different filter kernels are different from one another and different layers of the output tensor are calculated by different filter kernels.

Abstract: Methods and systems for detecting objects near an autonomous vehicle (AV) are disclosed. An AV will capture an image. A trained network will process the image at a lower resolution and generate a first feature map that classifies object(s) within the image. The system will crop the image and use the network to process the cropped section at a higher resolution to generate a second feature map that classifies object(s) that appear within the cropped section. The system will crop the first feature map to match a corresponding region of the cropped section of the image. The system will fuse the cropped first and second feature maps to generate a third feature map. The system may output the object classifications in the third feature map to an AV system, such as a motion planning system that will use the object classifications to plan a trajectory for the AV.

Abstract: The present disclosure provides a dynamic gesture identification method, a gesture interaction method and an interaction system. The interaction system includes: a dynamic vision sensor configured to trigger an event in accordance with movement of an object in a field of view relative to the dynamic vision sensor, and output an event data flow; a hand detection module configured to process the received event data flow to determine an initial position of a hand; a hand tracking module configured to determine a series of state vectors in accordance with the initial position of the hand; a gesture identification module configured to create an event cloud in accordance with event data corresponding to the obtained state vector, and process the event cloud to identify the gesture; and a command execution module configured to execute a corresponding operating command in accordance with the identified gesture.

Abstract: Systems and methods for constructing a panoramic view based on a changing viewpoint are disclosed. An exemplary system includes a storage device configured to receive first image data of a scene for a first viewpoint at a first predetermined elevation. The system further includes at least one processor configured to convert the first image data to candidate image data for at one or more second predetermined elevations using a deep learning neural network. The at least one processor is further configured to receive a user view request for virtually viewing the scene and determine second image data for a second viewpoint associated with the user view request, by mapping the second viewpoint to the one or more second predetermined elevations. The at least one processor is also configured to render the three-dimensional model of the scene based on the second image data and display the panoramic view in response to the user view request.

Abstract: A computer-implemented method for classifying video data with improved accuracy includes obtaining, by a computing system comprising one or more computing devices, video data comprising a plurality of video frames; extracting, by the computing system, a plurality of video tokens from the video data, the plurality of video tokens comprising a representation of spatiotemporal information in the video data; providing, by the computing system, the plurality of video tokens as input to a video understanding model, the video understanding model comprising a video transformer encoder model; and receiving, by the computing system, a classification output from the video understanding model.

Abstract: A digital video camera architecture for updating an object identification and tracking model deployed with the camera is disclosed. The invention comprises optics, a processor, a memory, and an artificial intelligence logic which may further comprise artificial neural networks. The architecture may identify objects according to a confidence threshold of a model. The confidence threshold may be monitored over time, and the model may be updated if the confidence threshold drops below an acceptable level. The data for retraining is ideally generated substantially internal to the camera. A filter is generated to process the entire field data set stored on the camera to create a field data subset also stored on the camera. The field data subset may be run through the model to generate cases that may be used in further monitoring, training, and updating of the model.

Abstract: In a method for defecting surface defects, a trained weighting generated when defect-free training samples are used to train an autoencoder and pixel convolutional neural network is obtained. A test encoding feature is obtained by inputting the trained weighting into the autoencoder and pixel convolutional neural network and a weighted autoencoder of the weighted autoencoder and pixel convolutional neural network encoding a test sample. The test encoding feature is input into a weighted pixel convolution neural network of the weighted autoencoder and pixel convolutional neural network to output a result of test. The test result is either no defect in the test sample or at least one defect in the test sample. Inaccurate determinations as to defects are thereby avoided. An electronic device and a non-transitory storage medium are also disclosed.

Abstract: A system comprises a computer including a processor and a memory. The memory includes instructions such that the processor is programmed to: determine, via a trained neural network model, a route for a vehicle to traverse based on vehicle sensor data, and update the trained neural network model based on data received from at least one of an edge computing device or an infrastructure (V2I) device.

Abstract: Generative AI systems and methods are provided to produce leading indicators of economic activity based on, for example, agricultural, fishing, mining, lumber harvesting, environmental, or ecological attributes and other factors determined from a range of available data sources. A consistent, semantic metadata structure is described as well as a hypothesis generating and testing system capable of generating predictive analytics models in a non-supervised or partially supervised mode. Users may then subscribe to the date for the use in economic forecasting.

Abstract: This application relates to an image data invoking method and system for an application, an electronic device, and a storage medium. When the electronic device identifies an entry file HalSensorList.cpp of a sensor, an image with an image resolution of binning size is filled. The electronic device obtains an image with an image resolution of binning size from a preview mode library at a kernel layer, fills an MTK_SCALER_AVAILABLE_STREAM_CONFIGURATIONS_WITH_DURATIONS tag at a Hal layer with the image with an image resolution of binning size, and reports the binning size of the MTK_SCALER_AVAILABLE_STREAM_CONFIGURATIONS_WITH_DURATIONS tag at the Hal layer to an android.scaler.availableStreamConfigurations tag at a framework layer. An application obtains the image with an image resolution of binning size with which the android.scaler.availableStreamConfigurations tag at the framework layer is filled from the android.scaler.availableStreamConfigurations tag.

Abstract: A system for training a multi-task model includes a processor and a memory in communication with the processor. The memory has a multi-task training module having instructions that, when executed by the processor, causes the processor to provide simulation training data having a plurality of samples to a multi-task model capable of performing at least a first task and a second task using at least one shared. The training module further causes the processor to determine a first value (gradience or loss) for the first task and a second value (gradience or loss) for a second task using the simulation training data and the at least one shared parameter, determine a task induced variance between the first value and the second value, and iteratively adjust the at least one shared parameter to reduce the task induced variance.

Abstract: Generative AI systems and methods are developed to provide recommendations regarding the prevention, detection, mitigation, and/or remediation of cybersecurity threats as determined from a range of available data sources. A consistent, semantic metadata structure is described as well as a hypothesis generating and testing system capable of generating predictive analytics models in a non-supervised or partially supervised mode. Users and/or AI agents (i.e., a form of “agentic AI”) may then subscribe to the data for the use in cybersecurity analytics, protection, mitigation, containment, remediation, and/or counterattacks of cybersecurity threats.

Abstract: A method includes accessing a web-based property over a network; storing a plurality of images or videos from the web-based property and associations between the plurality of images or videos and a target audience identifier responsive to the web-based property having a stored association with the target audience identifier; retrieving the plurality of images or videos from the database responsive to each of the plurality of images or videos having stored associations with the target audience identifier; executing a neural network to generate a performance score for each of the plurality of images or videos; calculating a target audience benchmark; executing the neural network to generate a first performance score for a first image or video and a second performance score for a second image or video; comparing the first performance score and the second performance score to the benchmark; and generating a record identifying the first image or video.

Abstract: Generative AI systems and methods are provided to provide recommendations as to whether a particular security associated with a corporate entity and/or its competitors should be purchased, sold, or held, as determined from a range of available data sources. A consistent, semantic metadata structure is described as well as a hypothesis generating and testing system capable of generating predictive analytics models in a non-supervised or partially supervised mode. Users may then subscribe to the date for the use in economic forecasting.

Abstract: A method and a system for detecting a scene text may include extracting a first feature map for a scene image input based on a convolutional neural network, and delivering the first feature map to a sequential deformation module; obtaining sampled feature maps corresponding to sampling positions by performing iterative sampling for the first feature map, obtaining a second feature map by performing a concatenation operation in deep learning according to a channel dimension for the first feature map and the sampled feature maps; obtaining a third feature map by performing a feature aggregation operation for the second feature map in the channel dimension, and delivering the third feature map to the object detection baseline network; and performing text area candidate box extraction for the third feature map and obtaining a text area prediction result as a scene text detection result through regression fitting.

Abstract: The present application discloses a method for filtering image feature points and a terminal. The method for filtering image feature points includes: providing quality score values to feature points extracted from an image, and according to the feature points and the quality score values of the feature points, training a neural-network model; after one time of filtering has started up, acquiring one frame of an original image and extracting feature points in the original image; inputting the original image and the feature points in the original image into the neural-network model, obtaining and outputting quality score values corresponding to the feature points in the original image; and according to the quality score values, filtering the feature points in the original image. The method for filtering image feature points can improve the success rate of the matching of the feature points in relocated application scenes, thereby improving the locating efficiency.

Abstract: A neural network compression method whereby forward inference is performed on target data by using a target parameter sharing network to obtain an output feature map of the last convolutional module, a channel related feature is extracted from the output feature map, the extracted channel related feature and a target constraint condition are input into a target meta-generative network, and an optimal network architecture under the target constraint condition is predicted by using the target meta-generative network to obtain a compressed neural network model.

Abstract: A coating analyzer is configured to receive electronic image data of a physical coating and to generate information regarding the pigments of the physical coating. The coating analyzer applies a computer vision model trained on baseline image data to the electronic image data. The coating analyzer assigns color values to the pigments forming the electronic image data and generates pigment groups based on the assigned color values. The pigment groups provide color palette data regarding the pigments forming the coating.

Abstract: An image processing method is provided. The image processing method includes: acquiring first second input images; extracting a content feature of the first input image; extracting an attribute feature of the second input image; performing feature fusion and mapping processing on the content feature of the first input image and the attribute feature of the second input image by using a feature transformation network to obtain a target image feature, the target image feature having the content feature of the first input image and the attribute feature of the second input image; and generating an output image based on the target image feature.