Alerta

FOCUSED HYPERPARAMETER TUNING USING ATTRIBUTION

Resumen de: US20260320412A1

0000 According to some embodiments, a method includes determining test predictions by performing an inference process using a checkpointed learning model and test vectors. The checkpointed learning model includes hyperparameters and weights. The method further includes determining an attribution map by performing one or more attribution processes using the test predictions and the test vectors. The method further includes determining a score for each particular hyperparameter by analyzing the attribution map using an association classifier. The method further includes determining, based on the analysis by the association classifier, whether each particular hyperparameter should be frozen or tuned again. The method further includes updating the hyperparameters and weights of the neural network when it is determined that at least one particular hyperparameter should be tuned again.

DEVICE WAKE-UP METHOD, SYSTEM, AND CHIP

Resumen de: US20260320806A1

0000 This application provides a device wake-up method, a system, and a chip. The method may include: processing a captured first image, determining a motion region image in the first image, and determining, by using a level-1 neural network, whether a target object exists in the motion region image; and when it is determined that the target object exists in the motion region image, performing reconfirmation, by using a level-2 neural network cascaded with the level-1 neural network, on a result output by the level-1 neural network. It can be learned that the target object is determined by using the level-1 neural network and the level-2 neural network, so that a false alarm frequency can be reduced, and system power consumption can be reduced.

Ventricular tachycardia exit site identification method and apparatus

Resumen de: AU2026201422A1

Abstract To non-invasively identify a ventricular tachycardia exit site in a heart of a particular patient (107) using electrocardiogram signals and medical imagery for the particular patient, a control circuit (101) can be configured to access (205) a neural network (104) that has been trained (202) using a plurality of synthetic data examples (203) and to input information regarding the electrocardiogram signals and the aforementioned medical imagery for the particular patient to the neural network. The control circuit can then output from the neural network information that identifies a predicted ventricular tachycardia exit site in the heart of the particular patient. Abstract eb b s t r a c t e b BY A CONTROL CIRCUIT GENERATE SYNTHETIC DATA EXAMPLES 201 SYNTHETIC TRAIN A DATA NEURAL DATA ACCESS A NEURAL NETWORK THAT HAS BEEN TRAINED INPUT INFORMATION REGARDING ELECTROCARDIOGRAM SIGNALS AND MEDICAL IMAGERY FOR A PARTICULAR 206 PATIENT TO THE NEURAL NETWORK OUTPUT FROM THE NEURAL NETWORK INFORMATION THAT IDENTIFIES A PREDICTED VENTRICULAR TACHYCARDIA 207EXIT SITE IN THE HEART OF THE PARTICULAR PATIENT eb e b -

APPLICATION PROGRAMMING INTERFACE TO INDICATE FRAME INTERPOLATION SUPPORT

Resumen de: US20260320741A1

Apparatuses, systems, and techniques to process image frames. In at least one embodiment, an application programming interface (API) is performed to indicate support to use one or more neural networks to perform frame interpolation.

NEURAL NETWORK-BASED NOISE SYNTHESIS

Resumen de: US20260320983A1

Apparatuses, systems, and techniques are presented to introduce noise or variation into images. In at least one embodiment, one or more neural networks are used to determine noise to be added to one or more images of one or more objects based, at least in part, upon one or more features of the one or more objects.

Prediction of bone based on point cloud

Resumen de: AU2026238109A1

A method for surgical planning includes obtaining, by a computing system, a first point cloud representing a first portion of a bone or a first cloud representing at least a portion of a bone, applying, by the computing system, a point cloud neural network to generate a second point cloud based on the first point cloud, the second point cloud comprising at least one of points representing at least a second portion of the bone or points representing an axis along the bone, and generating, by the computing system, surgical planning information based on the second point cloud. ep e p

ELECTRONIC APPARATUS AND CONTROLLING METHOD THEREOF

Resumen de: US20260315867A1

0000 An electronic apparatus is disclosed. The electronic apparatus includes: a memory including at least one instruction, a processor coupled with the memory and configured to control the electronic apparatus, and the processor is configured, by executing the at least one instruction, to: obtain a moving image, identify a person and pose data of the person from a plurality of frames in the moving image, obtain exercise pattern information corresponding to the plurality of frames using pose data of the identified person, recognize an exercise motion of the identified person by inputting at least one frame of the moving image into at least one neural network model, and obtain exercise feature information corresponding to the recognized exercise motion, identify, based on the exercise feature information and the exercise pattern information, a first frame interval and a second frame interval different from the first frame interval from among the plurality of frames in the moving image, and provide information on an exercise motion corresponding to the second frame interval by comparing the first frame interval with the second frame interval.

PLATING APPARATUS, AND METHOD OF ESTIMATING STATE IN PLATING APPARATUS

Resumen de: US20260317942A1

To provide a plating apparatus that makes it possible to estimate a current density at a plating-target surface with high precision even if a change in plating conditions occurs. The plating apparatus includes: a sensor configured to measure a potential of a plating solution in a vicinity of a peripheral portion of a substrate held by a substrate holder; and a state estimator configured to calculate, using the measured potential as input, a state estimate representing a current density at the peripheral portion of the substrate, based on an observation model and a state transition model. The state estimator is configured to execute computation based on the observation model by using a first trained neural network model. When a state variable representing the current density at the peripheral portion of the substrate is input, the first trained neural network model is configured to output an estimate representing the potential of the plating solution in the vicinity of the peripheral portion.

TECHNIQUES FOR CLASSIFICATION WITH NEURAL NETWORKS

Resumen de: US20260320156A1

0000 Apparatuses, systems, and techniques to train neural networks to perform classification. In at least one embodiment, one or more neural networks are trained to perform classification based, at least in part, on grouping one or more sets of neural network training data according to behaviors of one or more objects within one or more images represented by the training data.

INDEPENDENT POSITIONING OF AUXILIARY INFORMATION IN NEURAL NETWORK BASED PICTURE PROCESSING

Resumen de: US20260320995A1

Methods and apparatuses are provided for processing of picture data or picture feature data using a neural network with two or more layers. The present disclosure may be applied in the field of artificial intelligence (AI)-based video or picture compression technologies, and in particular, to the field of neural network-based video compression technologies. According to one or more embodiments, two kinds of data are combined during the processing by the neural network. The two kinds of data are obtained from different stages of processing by the network. Some of the advantages may include greater scalability and a more flexible design of the neural network architecture which may further lead to better encoding/decoding performance.

BRAKING ASSISTANCE METHOD AND SYSTEM BASED ON ROAD SURFACE RECOGNITION

Resumen de: WO2026214199A1

A braking assistance method and system based on road surface recognition. The braking assistance method comprises: S1, acquiring a road surface image, and processing the road surface image by means of a trained lightweight convolutional neural network, to obtain a road surface type, wherein the lightweight convolutional neural network introduces an inverted residual structure combined with a hybrid attention mechanism and introduces depthwise separable convolution; S2, determining road condition information by combining the road surface type and a preset database; S3, acquiring a rate of change in displacement of a brake pedal, and generating a braking intention of a driver on the basis of the rate of change in displacement of the brake pedal; and S4, dynamically adjusting the distribution of braking force on the basis of the road condition information and the braking intention of the driver, generating a braking control instruction, and sending the braking control instruction to an actuator. By means of the method, when a vehicle travels under adverse road conditions, the vehicle speed can be controlled to maintain vehicle stability, thereby solving the problem of poor dynamic stability in existing vehicle braking assistance.

SYSTEM AND METHOD FOR MICROMECHANICS SIMULATION

Resumen de: US20260320267A1

A micromechanics simulation framework has the steps of collecting one or more images of at least one composite material, providing, to a first neural network, an input including the one or more images, segmenting one or more microstructures in the one or more images, quantifying one or more physical descriptors of one or more segmented microstructures, passing an output including the one or more segmented microstructures and one or more quantified physical descriptors to one of: a statistical-based module using descriptor-driven stochastic construction routines with periodic boundary conditions, or a second deep generative neural network, and generating one or more stochastic representative volume element (SRVE) ensembles based on the one or more segmented microstructures and one or more quantified physical descriptors. A micromechanics simulation system is also disclosed for computing the homogenized elastic properties and the time-dependent response of composites in high-temperature environments.

ATTRIBUTE-BASED ENCRYPTION FOR SELECTIVE DOCUMENT CONTENT PROTECTION

Resumen de: EP4823042A2

0001 The present disclosure provides a method for securing document content. A document security manager receives a digital document containing original text and image content. The original text content is processed to identify text portions by parsing annotations, analyzing proximate text, extracting content within the determined scope, and determining encryption attributes. The original image content is processed to identify image portions by detecting user-defined selections or employing a neural network to identify sensitive objects. Encrypt content requests are generated including the identified text and image portions with associated attributes and metadata. The requests are transmitted to a cryptographic engine, which returns encrypted content. A partially encrypted document is built by replacing original content with masking symbols, applying visual obfuscation, embedding encrypted content as metadata, and updating document properties to indicate encrypted content.

METHOD FOR GENERATING INTERNAL STRUCTURE OF ORGANISM ON BASIS OF DEEP LEARNING

Resumen de: EP4822642A1

The present application relates to the field of life sciences and medical imaging device technologies, and more particularly, to a deep learning-based method for generating an internal structure of an organism. The method includes the following steps: S1, constructing and training an internal organ generation model of a imaging target; S2, obtaining a three-dimensional body surface contour image of the imaging target; S3, inputting the obtained three-dimensional body surface contour image of the imaging target into the internal organ generation model to obtain a three-dimensional internal organ distribution image; and S4, performing merging on the three-dimensional internal organ distribution image and the three-dimensional body surface contour image, and displaying a result obtained based on the merging. With a deep neural network-based internal organ generation model of the imaging target, the three-dimensional internal organ distribution image of the imaging target can be obtained quickly and accurately simply through obtaining the three-dimensional body surface contour image of the imaging target and inputting the three-dimensional body surface contour image into the internal organ generation model, which simplifies operation steps, realizes automatic imaging, improves an imaging efficiency, and has high applicability.

Neural networks for indicating priority of connection requests

Resumen de: GB2705387A

Apparatuses, systems, and techniques are described to predict priority of a connection request (e.g., a next connection request). For example, processing circuitry uses one or more neural networks to indicate priority of one or more next connection requests within one or more radio access network (RAN) networks. A base station of a UE device can use, perform, or otherwise execute the one or more neural networks to generate a prediction of what a next connection request will be. Fig. 1

PANORAMA GENERATION USING ONE OR MORE NEURAL NETWORKS

Resumen de: US20260311290A1

0000 Apparatuses, systems, and techniques are presented to generate panoramas from individual images. In at least one embodiment, one or more generative neural networks are used to generate a spherical panoramic image using features extracted from a single input image.

Identification of multi-scale features using neural network

Resumen de: AU2026234120A1

Apparatuses, systems, and techniques to identify features within one or more images. In at least one embodiment, features are identified in one or more images using one or more neural networks containing convolutional layers with multiple filters that may be executed by one or more parallel processing unit. ep e p

TRAINING VISION LANGUAGE NEURAL NETWORKS WITH COMPONENT REUSE

Resumen de: US20260311476A1

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating and training a glued neural network that reuses components taken from a first pre-trained neural network and a second pre-trained neural network. For example, the first pre-trained neural network can be a multi-modal neural network that includes a visual encoder neural network and a small-scale language model neural network. The second pre-trained neural network can be a large-scale language model neural network.

Systems and Methods for Direct-From-Non-Volatile-Memory Neural Network Inference and Hidden-State Reuse

Resumen de: US20260310378A1

0000 Systems and methods for performing inference with a neural network model using model weights stored in non-volatile memory without loading the model weights into dynamic random-access memory (DRAM) prior to inference. A processing system reads model weights from non-volatile memory, streams the model weights to a compute engine, generates outputs by performing multiply-accumulate operations using the streamed model weights, and maintains intermediate states in on-chip volatile memory. In some embodiments, a recurrent computational model maintains a hidden state with fixed dimensionality independent of sequence length. In some embodiments, the processing system pre-processes context data to generate hidden states, stores the hidden states, and retrieves the hidden states to resume inference without reprocessing the context data. The hidden states may be stored as tuples comprising a key and a hidden state in a vector database.

DYNAMIC PARALLELISM

Resumen de: US20260311000A1

0000 In various examples, one or more neural networks are to be performed by a number of processors. The neural network may be reconfigured during inferencing such that the number of processors that start the inferencing is different from the number of processors that finish the inferencing.

TEMPORAL FEATURE SCALING FOR NEURAL NETWORK PROCESSING

Resumen de: US20260311361A1

Systems, devices, and methods described herein relate to feature scaling for neural network processing. In some examples, temporal feature scaling is employed for efficient neural network processing. To process the neural network, a processing system accesses feature maps of a current input frame in a temporal sequence of input frames and scales the feature maps of the current input frame using excitation values generated from a previous input frame in the temporal sequence. The processing system generates new excitation values from the current input frame to perform scaling for a subsequent input frame in the temporal sequence.

High Likeness High Compliance Human Figure Generation

Resumen de: US20260311382A1

Techniques relating to high likeness high compliance human figure generation are disclosed. A method for high likeness high compliance human figure generation may include selecting training images as a basis for an expert learning model, obtaining embeddings for the training images, the embedding being recorded separately for each expert, computing a geometric property of the embedding space for each expert, embedding a received prompt using the same embedding method as for the training images, and generating an output image using a subset of experts selected for highest compliance (e.g., most similarity) to the received prompt, or for other purposes. A system for implementing this method may include a training data ensemble, an image augmentation pipeline, an LLM and other neural networks, a mood and style ensemble of lightweight expert models built on a foundational model, and a tokenizer.

MISSION PREDICTION SYSTEM FOR SIMULTANEOUS MISSION SUCCESS AND ENEMY INTENT PREDICTION USING GRAPH NEURAL NETWORKS

Resumen de: US20260310955A1

The mission prediction system models a wargaming scenario as a graph. Once the GNN graph is defined, multiple sets of AIGC data is used to train the GNN graph. The trained GNN graph can then be used to perform inference on novel wargaming scenarios to simultaneously predict mission success and enemy intent in real-time, even if the novel wargaming scenarios involve a much greater number of assets than were used in the AIGC data. A viewer can be used to review the predictions and to analyze the novel wargaming scenario at different timesteps.

Corrosion Type Identification in Flanged Joints Using a Novel Recurrent Neural Network Technique Applied to Electrochemical Noise Measurements

Resumen de: US20260309616A1

A method is provided for automatically determining corrosion-related temporal features using an autoencoder that considers and has as an input electrochemical noise (EN) data from a metallic structure, wherein the determined corrosion-related temporal features are used to classify and/or characterize types of corrosion of the metallic structure. The method relies on the autoencoder and EN data to automatically determine the corrosion-related temporal features without requiring prior labeling of data or human supervision. The disclosed method may be implemented in a computer program product or an apparatus that enable classification or characterization of types of corrosion of the metallic structure. In some embodiments, the method may include the step of measuring the raw EN data and the apparatus may include an EN measurement device for measuring raw EN data.

DEEPFAKE SHIELDING OF IMAGES

Nº publicación: WO2026213578A1 08/10/2026

Solicitante:

STEG AI CORP [US]
STEG AI CORPORATION

Resumen de: WO2026213578A1

Embodiments of the present disclosure include techniques for processing images that depict human faces. In one embodiment, one or more source images comprising a source identity are received. At runtime, a source image is applied to a neural network to generate adversarial pixel signals to be applied to the source image. The source image to generate a shielded image that retains the source identity but with modified pixels depicting the face of the person. The shielded image is provided by the image processing system. Applying the shielded image to a face swapper produces a swap identity that is visually distinct from the source identity. The visual distinctions between the swap identity and the source identity are caused by an adversarial attack that emerges in a facial detection component, a landmark detection component, and/or an identity mapping component of the face swapper.