In various examples, techniques for augmenting object classification using metadata associated with objects are described herein. Systems and methods described herein may process metadata associated with objects along with sensor data representing the objects when performing object classification. For instance, if the sensor data includes image data, a bounding shape (e.g., a bounding box, etc.) associated with an object may be used to generate a cropped image of the object. The metadata associated with the object may then be determined, where the metadata may represent information associated with a geographic area for which the object is located, information associated with the bounding shape (e.g., coordinates, dimensions, an aspect ratio, etc.), and/or any other information. One or more machine learning models may then process input representing the cropped image along with the metadata to determine a classification associated with the object.
G06V 10/26 - Segmentation de formes dans le champ d’imageDécoupage ou fusion d’éléments d’image visant à établir la région de motif, p. ex. techniques de regroupementDétection d’occlusion
G06V 10/764 - Dispositions pour la reconnaissance ou la compréhension d’images ou de vidéos utilisant la reconnaissance de formes ou l’apprentissage automatique utilisant la classification, p. ex. des objets vidéo
G06V 10/80 - Fusion, c.-à-d. combinaison des données de diverses sources au niveau du capteur, du prétraitement, de l’extraction des caractéristiques ou de la classification
G06V 20/58 - Reconnaissance d’objets en mouvement ou d’obstacles, p. ex. véhicules ou piétonsReconnaissance des objets de la circulation, p. ex. signalisation routière, feux de signalisation ou routes
2.
APPLICATION PROGRAMMING INTERFACE TO INDICATE PROCESSOR ACTIVITY
Apparatuses, systems, and techniques to identify a clock frequency at which one or more processors are to operate. In at least one embodiment, a processor performs an application programming interface (API) to cause one or more activity levels of one or more processors to be indicated to one or more users.
Apparatuses, systems, and techniques to identify a clock frequency at which one or more processors are to operate. In at least one embodiment, a processor performs an application programming interface (API) to cause one or more one or more activity levels of one or more processors to be measured at one or more indicated intervals.
Apparatuses, systems, and techniques to identify a clock frequency at which one or more processors are to operate. In at least one embodiment, a processor performs an application programming interface (API) to cause one or more statistics corresponding to one or more activity levels of one or more processors to be indicated to one or more users.
Systems and methods herein are for leak detection in a computing environment using a shaped leak sensor. The shaped leak sensor may include an insulating material, printed or applied thereon conductive traces on a first side, and an adhesive on a second side. The shaped leak sensor may be configured to be shaped to match at least a layout around a component in the computing environment. A detector can monitor an input from the shaped leak sensor to determine one or more of different states associated with the shaped leak sensor.
G01M 3/16 - Examen de l'étanchéité des structures ou ouvrages vis-à-vis d'un fluide par utilisation d'un fluide ou en faisant le vide par détection de la présence du fluide à l'emplacement de la fuite en utilisant des moyens de détection électrique
H05K 7/20 - Modifications en vue de faciliter la réfrigération, l'aération ou le chauffage
A textual transcript and one or more language indicators are determined using a multilingual speech-to-text (STT) model of a multilingual automatic speech recognition (ASR) system and using an audio sample as input to the multilingual STT model. The textual transcript is associated with the audio sample, and the one or more language indicators are each associated with a respective grammatical unit of one or more grammatical units of the textual transcript. A monolingual language model (LM) of a plurality of monolingual LMs of the ASR system is identified using a language indicator of the one or more language indicators. The textual transcript associated with the audio sample is caused to be refined using the identified LM and using a subset of the textual transcript as input to the identified LM.
G10L 15/19 - Contexte grammatical, p. ex. désambiguïsation des hypothèses de reconnaissance par application des règles de séquence de mots
G06F 40/58 - Utilisation de traduction automatisée, p. ex. pour recherches multilingues, pour fournir aux dispositifs clients une traduction effectuée par le serveur ou pour la traduction en temps réel
G10L 15/06 - Création de gabarits de référenceEntraînement des systèmes de reconnaissance de la parole, p. ex. adaptation aux caractéristiques de la voix du locuteur
G10L 15/187 - Contexte phonémique, p. ex. règles de prononciation, contraintes phonotactiques ou n-grammes de phonèmes
G10L 15/30 - Reconnaissance distribuée, p. ex. dans les systèmes client-serveur, pour les applications en téléphonie mobile ou réseaux
A pose of the calibration target that is associated with the image is determined determining for each image of the one or more images. The pose includes at least one of a position or an orientation of the calibration target in a local coordinate system of a local positioning system. The image capture device is calibrated based on determining a relationship between the position of the calibration target in the one or more images and the associated local pose of the calibration target in the local coordinate system.
Processors, systems and techniques to encode an input data set as a sequence of encoded values are described. In at least one embodiment, an input data set is obtained and encoded using one or more neural networks as a sequence of encoded values based, at least in part, on similarity measurements between encoded values in the sequence.
A technique for reconstructing a three-dimensional scene from monocular video adaptively allocates an explicit sparse-dense voxel grid with dense voxel blocks around surfaces in the scene and sparse voxel blocks further from the surfaces. In contrast to conventional systems, the two-level voxel grid can be efficiently queried and sampled. In an embodiment, the scene surface geometry is represented as a signed distance field (SDF). Representation of the scene surface geometry can be extended to multi-modal data such as semantic labels and color. Because properties stored in the sparse-dense voxel grid structure are differentiable, the scene surface geometry can be optimized via differentiable volume rendering.
Systems and methods implement a technique for altering the shape of the cells by shifting coordinates of points along cell boundaries using a set of periodic functions. To avoid having cell boundaries along the scene surfaces, wavelengths of those periodic functions are selected so they are not a multiple of an original discretization. The coordinates may be shifted along different axes of the cells and may generate different cells having a variety of different outlines to reduce a likelihood of a cell boundary being positioned along a scene boundary.
G06T 19/00 - Transformation de modèles ou d'images tridimensionnels [3D] pour infographie
G06F 30/20 - Optimisation, vérification ou simulation de l’objet conçu
G06T 7/586 - Récupération de la profondeur ou de la forme à partir de plusieurs images à partir de plusieurs sources de lumière, p. ex. stéréophotométrie
11.
FAULT-TRIGGERED CHECKPOINTING FOR DISTRIBUTED TRAINING OF ARTIFICIAL INTELLIGENCE MODELS
In various examples, systems and techniques are provided that are directed to fault-triggered checkpointing of parallel training of machine learning models. Training involves a plurality of iterations including one or more fault-free iterations and a faulty iteration. The fault-free iterations include receiving, from an individual training process of a plurality of parallel training processes, a reference signal. The reference signal is associated with completion, by the individual training process, of the individual iteration. The faulty iteration includes retrieving, responsive to determining that no reference signal has been received from one or more training processes, a state of training of the model from a memory device associated with a fault-free training process. The state of training is then used to cause the training of the model to be restarted.
In various examples, localized noise reduction adaptation for color and infrared data channel processing is provided. Embodiments provide systems and methods for an ISP pipeline that address noise components introduced into RGB color channels due to adaptive adjustments to RGB color channels, such as local adaptation-based IR subtraction adjustments. Cumulative noise gain information may be communicated in the form of an adaptive noise gain map. A noise model adjustment function may use correction information from the adaptive noise gain map to dynamically compute supplemental noise adjustments that represent noise corrections relative to a sensor noise profile used by a noise reduction stage of the ISP for noise correction. Application of the supplemental noise adjustments to the sensor noise profile may be represented as a composite noise map that is input to the noise reduction stage.
Disclosed are systems and techniques for AI stereo disparity estimation. The techniques include generating a cost volume matrix based on a stereo image pair. The techniques include generating a disparity maps for the first image of the stereo image pair, which includes, for each pixel in the first image, generating a disparity value corresponding to the pixel by performing stereo image processing on the cost volume matrix entry corresponding to the pixel to generate an intermediate stereo image processing output, generating, using the intermediate stereo image processing output as input to a CNN, one or more weight values, and calculating, for the pixel, the disparity value using one or more intermediate disparity values of the intermediate stereo image processing output and the plurality of weight values.
Feedforward reasoning models that include a video encoder configured to generate feature tokens from an input video, logic to condition the feature tokens with camera parameters, at least one sparse attention head with two-way attention to transform settings from the feature tokens into a tracking token, a depth token, and a visibility token in accordance with an input prompt, and logic configured to transform the tracking token, depth token, and visibility token into track predictions for an object specified by the input prompt.
G06V 10/771 - Sélection de caractéristiques, p. ex. sélection des caractéristiques représentatives à partir d’un espace multidimensionnel de caractéristiques
G06V 10/80 - Fusion, c.-à-d. combinaison des données de diverses sources au niveau du capteur, du prétraitement, de l’extraction des caractéristiques ou de la classification
Apparatuses, systems, and techniques to perform operations in a processor asynchronously. In at least one embodiment, a processor performs perform at least one tensor instruction concurrently with one or more other instructions based, at least in part, on one or more indicators of the tensor instruction being asynchronous.
Embodiments of the present disclosure relate to a system and method used to transfer image data via Ethernet. The system may include memory for storing frame data that may be received via Ethernet packets. In particular, the Ethernet packets may include a payload that may include one or more segments and a header. The header may include a sequence number field indicating a respective sequence number that corresponds to the respective segment, and a byte offset field that may indicate a respective byte offset that may be applied to the segment. Further, the system may include hardware that may be configured to perform packet analysis operations including determining whether a previously transmitted segment was lost. The system may additionally include a processing system for performing data processing operations including storing individual segments at respective memory locations based on the respective byte offsets included in the Ethernet packets.
Apparatuses, systems, and techniques to predict object trajectories through areas of an environment that are occluded from cameras. In at least one embodiment, a trajectory of an object through an area occluded from cameras is predicted using one or more neural networks, based on, for example, known positions of the object in the environment and information describing a stationary object in the environment.
G06V 20/58 - Reconnaissance d’objets en mouvement ou d’obstacles, p. ex. véhicules ou piétonsReconnaissance des objets de la circulation, p. ex. signalisation routière, feux de signalisation ou routes
G06V 10/25 - Détermination d’une région d’intérêt [ROI] ou d’un volume d’intérêt [VOI]
G06V 10/82 - Dispositions pour la reconnaissance ou la compréhension d’images ou de vidéos utilisant la reconnaissance de formes ou l’apprentissage automatique utilisant les réseaux neuronaux
18.
GENERATING IMAGES OF VIRTUAL ENVIRONMENTS USING ONE OR MORE NEURAL NETWORKS
Apparatuses, systems, and techniques are presented to generate images. In at least one embodiment, one or more neural networks are used to generate one or more images based, at least in part, upon one or more semantic features projected from a three-dimensional environment.
The disclosed method for training one or more robot control models includes performing, based on one or more demonstration trajectories of a robot performing one or more skills associated with a task, one or more training operations to generate one or more first trained machine learning models for controlling the robot; and performing one or more reinforcement learning operations using the one or more first trained machine learning models to generate one or more second trained machine learning models for controlling the robot.
Apparatuses, systems, and techniques to identify reasons of users canceling a subscription to an online service. In at least one embodiment, one or more reasons one or more users stop using an online services are identified using one or more neural networks, based on, one or more representative reasons among a plurality of reasons one or more users stopped using said online service. In addition, apparatuses, systems, and techniques to identify reasons of users canceling a subscription to an online service. In at least one embodiment, one or more reasons one or more users stop using an online services are identified using one or more neural networks, based on, for example, one or more interactions with said online service by one or more users.
G06Q 30/0282 - Notation ou évaluation d’opérateurs commerciaux ou de produits
G06Q 30/016 - Fourniture d’une assistance aux clients, p. ex. pour assister un client dans un lieu commercial ou par un service d’assistance après-vente
21.
TECHNIQUES FOR GENERATING MOLECULES WITH FRAGMENT RETRIEVAL AUGMENTATION
The disclosed method for generating molecules includes selecting, based on one or more molecule properties, one or more hard molecule fragments and one or more soft molecule fragments; and processing, using a trained machine learning model, the one or more hard molecule fragments and the one or more soft molecule fragments to generate a molecule, where the molecule includes the one or more hard molecule fragments, and the trained machine learning model generates the molecule based on the one or more soft molecule fragments.
G16B 15/00 - TIC spécialement adaptées à l’analyse de structures moléculaires bidimensionnelles ou tridimensionnelles, p. ex. relations structurelles ou fonctionnelles ou alignement de structures
According to various embodiments, a processing subsystem includes a housing; a printed circuit board (PCB) disposed within the housing; an integrated circuit package that has a first side and a second side that is opposite to the first side, wherein the first side of the integrated circuit package is mounted on the PCB; and a liquid-based cooling system. The liquid-based cooling system is disposed within the housing and includes: at least one radiator element; a pump that is fluidly coupled to the radiator element; and at least one fan that directs cooling air across the radiator element.
Calibration of various sensors may be difficult without specialized software to process intrinsic and extrinsic information about the sensors. Certain types of input files, such as image files, may also lack certain information, like depth information, to effectively translate regions of interest between images taken from a different perspective. Landmarks can be used to establish points for associating regions of interest between images taken from a different perspective and provided as an overlay to verify sensor calibration.
Techniques for emergent scene decomposition from multi-traverse include receiving a plurality of images from multiple traversals of a scene; generating a plurality of 3D Gaussians from the plurality of images; projecting each of the plurality of 3D Gaussians to generate a plurality of rendered 2D images; extracting a feature map from each of the plurality of images and the plurality of rendered 2D images; generating ephemeral objects masks for the plurality of images from the feature maps and the plurality of rendered 2D images; generating optimized 3D Gaussians from the plurality of images, the plurality of rendered 2D images, and the ephemeral objects masks; and generating a 3D environment from the optimized 3D Gaussians.
Apparatuses, systems, and techniques to use different resolutions of data as part of inferencing. In at least one embodiment, embeddings corresponding to different resolutions of data to be encoded by the embeddings may be obtained in order to be used as part of one or more neural networks that include the embeddings.
Apparatuses, systems, and techniques to cause one or more non-unique wireless device identifiers to correspond to one or more unique wireless device identifiers. In at least one embodiment, one or more non-unique wireless device identifiers are mapped to one or more unique wireless device identifiers.
A system includes a memory device and one or more processing devices operatively coupled to the memory device via a memory channel. The processing device(s) cause data to be received over the memory channel from the memory device, where the data includes known multi-bit patterns. The processing device(s) sweep the data over voltage and time to generate eye diagram data. The processing device(s) detect errors at identified cursors of the eye diagram data, where each identified cursor corresponds to a known multi-bit pattern within a set of previously transmitted bits. The processing device(s) store counts of each detected error associated with a respective known multi-bit pattern. The processing device(s) determine, using the counts, a plurality of decision feedback equalizer (DFE) coefficients to be employed in receiving unknown data over the memory channel.
H04L 1/00 - Dispositions pour détecter ou empêcher les erreurs dans l'information reçue
H03M 13/29 - Codage, décodage ou conversion de code pour détecter ou corriger des erreursHypothèses de base sur la théorie du codageLimites de codageMéthodes d'évaluation de la probabilité d'erreurModèles de canauxSimulation ou test des codes combinant plusieurs codes ou structures de codes, p. ex. codes de produits, codes de produits généralisés, codes concaténés, codes interne et externe
H03M 13/39 - Estimation de séquence, c.-à-d. utilisant des méthodes statistiques pour la reconstitution des codes originaux
28.
APPLICATION PROGRAMMING INTERFACE TO CAUSE MEASUREMENT OF PROCESSOR ACTIVITY
Apparatuses, systems, and techniques to identify a clock frequency at which one or more processors are to operate. In at least one embodiment, a processor performs an application programming interface (API) to cause one or more measurements of one or more activity levels of one or more processors to be stopped.
Light contribution information can be determined and cached for use in rendering image frames for a scene. In at least one embodiment, a spatial hash data structure can be used to split the scene into regions, such as octahedral voxels. Using cast light rays, an average light contribution can be computed for each individual voxel. Those light values can then be used to build a cumulative distribution function for each voxel that can be used to select which lights to sample for a given frame during rendering. The sampling for a region or voxel can be based at least in part upon the number of contributing lights for that region, as well as the relative contributions of those lights. Such an approach can be very bandwidth and cache efficient, while providing high image quality.
Apparatuses, systems, and techniques to estimate exponent values to be used in data type conversions. In at least one embodiment, one or more largest exponent values of a plurality of floating point operands are identified based, at least in part, on calculating one or more probabilities of less than all of the plurality of floating point operands having the one or more largest exponent values.
Apparatuses, systems, and methods to perform thread memory addressing. In at least one embodiment, a processor comprising one or more circuits to perform an application programming interface (API) to generate one or more addresses for one or more instructions corresponding to a same software kernel.
Processors, systems and techniques to dynamically configure multiple variants of neural networks, including enabling or disabling portions of the neural networks, are disclosed. In at least one embodiment, portions of the neural networks may be enabled or disabled based at least in part on user configurable indicators of performance characteristics of processors or systems hosting the neural networks.
Triplanes are data representations used in computer graphics to encode scenes into compact feature representations that balance expressiveness with efficiency. Despite their efficiency, triplanes still suffer from large data bandwidth size, precluding use in streaming or dynamic settings. Methods which aim to compress triplanes, however, must be trained alongside the model and as a result are not generalizable among different scenes. The present disclosure provides a generalizable solution for triplane compression that can be applied to various triplanes without scene-specific training or finetuning.
H04N 19/597 - Procédés ou dispositions pour le codage, le décodage, la compression ou la décompression de signaux vidéo numériques utilisant le codage prédictif spécialement adapté pour l’encodage de séquences vidéo multi-vues
Apparatuses, systems, and techniques to select a neural network to perform one or more wireless signal operations, such as channel estimation. In at least one embodiment, a processor comprising one or more circuits is to use one or more signal-to-noise (SNR) values to select one or more neural networks to generate one or more channel estimates.
In various examples, systems and methods are disclosed related to performing generation of meeting insights for a live meeting using artificial intelligence (AI) technology. Such meeting insights may be in the form of inquiry responses that respond to an inquiry asked in a live meeting and/or orchestration directives that provide instruction or guidance for managing a flow of a live meeting. In embodiments, various data may be searched to identify contextual data for use in generating a meeting insight. For example, live meeting data, prior meeting data, and/or organizational data may be searched to identify contextual data relevant to a management event triggering generation of a meeting insight. The contextual data may then be used as input to a machine learning model, such as an LLM, VLM, or MMLM, to generate a corresponding meeting insight.
Apparatuses, systems, and techniques to cause, use, and/or perform, one or more neural networks, such as to locate an object. In at least one embodiment, one or more processors comprising one or more circuits is to use one or more neural networks to identify one or more probabilities of one or more ranges of depth of one or more pixels of one or more objects in relation to one or more cameras.
A transmitter device includes transmitter logic coupled to control logic, the control logic to receive data to be sent via a communication network, determine whether a first portion of the data matches a first data pattern, identify a first index corresponding to the first portion of the data, generate metadata for the data based on the first index, generate compressed data by removing the first portion of the data from the first index of the data, generate a compressed data signal based on the compressed data and the metadata, and cause the compressed data signal to be transmitted via the communication network.
In various examples, Automatic Test Equipment (ATE) tests may be applied to electronic components, or devices under test (DUTs), using high-speed functional interfaces. For instance, a DUT and an ATE host device may be configured to perform one or more handshake mechanisms to optimize ATE test flows. The handshakes may be used to pause and/or resume a test interface executing on the DUT. In some examples, the ATE host device and the DUT may use one or more common data storage locations for performing the handshakes, such as system memory of the ATE host device, addressable register space of the DUT, a combination thereof, or any other shareable memory space. Additionally, the test interface of the DUT may be configured to convert packetized test data received using the high-speed functional interface to a raw data format consistent with traditional ATE test inputs received using I/O pins of the DUT.
Methods and systems are disclosed for estimating 3D poses and sizes of vehicle occupants using neural networks. An image of the vehicle's interior is captured and a monocular depth map is generated. Both the depth map and the image are fed into a 3D pose estimation network as a combined four-channel RGBD input. The neural network may include an occlusion-aware masking layer that generates occlusion scores for key points associated with the occupant. The occlusion scores help the network adjust the weighting of depth information, such that key points with higher occlusion scores, indicating a greater likelihood of being hidden or partially obscured, receive lower weight. Scaling functions that estimate scale factors integrate depth information with the occlusion-aware masks to estimate the absolute depth positions of key points.
G06V 20/59 - Contexte ou environnement de l’image à l’intérieur d’un véhicule, p. ex. concernant l’occupation des sièges, l’état du conducteur ou les conditions de l’éclairage intérieur
G06T 7/246 - Analyse du mouvement utilisant des procédés basés sur les caractéristiques, p. ex. le suivi des coins ou des segments
G06T 7/50 - Récupération de la profondeur ou de la forme
G06V 10/25 - Détermination d’une région d’intérêt [ROI] ou d’un volume d’intérêt [VOI]
G06V 10/26 - Segmentation de formes dans le champ d’imageDécoupage ou fusion d’éléments d’image visant à établir la région de motif, p. ex. techniques de regroupementDétection d’occlusion
G06V 10/82 - Dispositions pour la reconnaissance ou la compréhension d’images ou de vidéos utilisant la reconnaissance de formes ou l’apprentissage automatique utilisant les réseaux neuronaux
G06V 20/70 - Étiquetage du contenu de scène, p. ex. en tirant des représentations syntaxiques ou sémantiques
Disclosed are circuits and techniques for efficient execution of dependent tasks. The techniques include retrieving from a memory a consumer task that depends on a producer task executed by a processing device and retrieving from the memory a first other task. The techniques further include, responsive to a first execution state of the producer task at a first time satisfying a first execution criterion, providing the first other task for execution by the processing device. The techniques further include, responsive to a second execution state of the producer task at a second time satisfying a second execution criterion, providing the consumer task for execution by the processing device.
Systems and methods described relate to the generation of image content. In order to provide for smoothing between sequential images, but avoid introducing lag into lighting effects, light information can be compared for regions between consecutive rendered frames. Shading can be performed and the results compared for tiles of pixels to compute gradient values, such as by using a single light sample for each tile. A filtering pass can be performed with respect to these gradients, and this filtered, lower-resolution grid version can be upscaled into a full resolution, screen-sized image and the gradients transformed into confidence values. These confidence values can be used to determine an extent to which to keep lighting data from the previous frame with respect to the current frame. For example, less lighting information can be used from the prior frame for a given pixel location if the confidence for that location is lower.
G06T 3/4053 - Changement d'échelle d’images complètes ou de parties d’image, p. ex. agrandissement ou rétrécissement basé sur la super-résolution, c.-à-d. où la résolution de l’image obtenue est plus élevée que la résolution du capteur
G06T 5/20 - Amélioration ou restauration d'image utilisant des opérateurs locaux
Apparatuses, systems, and techniques to identify reasons of users canceling a subscription to an online service. In at least one embodiment, one or more reasons one or more users stop using an online services are identified using one or more neural networks, based on, one or more representative reasons among a plurality of reasons one or more users stopped using said online service. In addition, apparatuses, systems, and techniques to identify reasons of users canceling a subscription to an online service. In at least one embodiment, one or more reasons one or more users stop using an online services are identified using one or more neural networks, based on, for example, one or more interactions with said online service by one or more users.
Systems and methods for atomic memory operations in a remote direct memory access network are disclosed. A system includes a network interface card (NIC) comprising a first memory and one or more processors coupled to the first memory. The one or more processors are to receive an atomic memory operation (AMO) remote procedure call (RPC) comprising a memory address and an AMO type. The one or more processors are further to retrieve a value corresponding to the memory address of the AMO RPC from a second memory. The one or more processors are further to perform an AMO corresponding to the AMO type on the value from the second memory to obtain a modified value. The one or more processors are further to store the modified value in the first memory.
In various examples, a technique for removing residual IR estimated locally at a pixel in an image and performing corresponding color correction operations locally at the same pixel is disclosed. The technique includes receiving a first image captured using an image sensor. The technique also includes removing, from a first pixel in the first image, a contribution of infrared radiation to a color channel of the first pixel based at least on an estimated amount of residual infrared radiation. The technique also includes performing a first part of local color correction for the color channel of the first pixel based at least on the estimated amount of residual infrared radiation. The technique also includes performing global color correction for a plurality of pixels in a second image that is reconstructed from the first image, where the global color correction includes a second part of the local color correction.
H04N 23/11 - Caméras ou modules de caméras comprenant des capteurs d'images électroniquesLeur commande pour générer des signaux d'image à partir de différentes longueurs d'onde pour générer des signaux d'image à partir de longueurs d'onde de lumière visible et infrarouge
G06T 7/90 - Détermination de caractéristiques de couleur
H04N 9/67 - Circuits pour le traitement de signaux de couleur pour le matriçage
H04N 9/73 - Circuits pour l'équilibrage des couleurs, p. ex. circuits pour équilibrer le blanc ou commande de la température de couleur
45.
3D GAUSSIAN FEATURE OPTIMIZATION BY DISTILLATION FROM 2D FOUNDATION MODELS
Generalizable feature distillation systems that align 3D features with 2D foundation model features using a feedforward network, avoiding per-scene optimization, and a flexible end-to-end 3D scene interpretation system that applies the extracted 3D features and pretrained 2D vision-language models for various 3D scene understanding tasks.
G06V 10/82 - Dispositions pour la reconnaissance ou la compréhension d’images ou de vidéos utilisant la reconnaissance de formes ou l’apprentissage automatique utilisant les réseaux neuronaux
G06V 10/74 - Appariement de motifs d’image ou de vidéoMesures de proximité dans les espaces de caractéristiques
G06V 10/77 - Traitement des caractéristiques d’images ou de vidéos dans les espaces de caractéristiquesDispositions pour la reconnaissance ou la compréhension d’images ou de vidéos utilisant la reconnaissance de formes ou l’apprentissage automatique utilisant l’intégration et la réduction de données, p. ex. analyse en composantes principales [PCA] ou analyse en composantes indépendantes [ ICA] ou cartes auto-organisatrices [SOM]Séparation aveugle de source
46.
USING STABLE DIFFUSION TO GENERATE SEAMLESS CONTENT TILE SETS IN CONTENT GENERATION SYSTEMS AND APPLICATIONS
Approaches presented herein can utilize a network that learns to generate a set of content tiles that represent a type of content (e.g., texture) and satisfy a set of rules or boundary conditions. The network can be a diffusion network that learns or adapts to the boundary conditions over several iterations. An indication of a type of content, along with a set of noisy prior images, can then be provided as input to the trained diffusion network, which can generate a set of content images. The content images can then be placed using a random (or other) selection process, as long as each selection satisfies the respective boundary conditions. Such an approach enables a small number of content tiles to be used for a texture region with a repeatability or pattern that may not be obviously detectable by a typical human viewer.
G06V 10/82 - Dispositions pour la reconnaissance ou la compréhension d’images ou de vidéos utilisant la reconnaissance de formes ou l’apprentissage automatique utilisant les réseaux neuronaux
Feedforward reasoning models that include a video encoder configured to generate feature tokens from an input video, at least one dense attention head, at least one sparse attention head with two-way attention logic configured to transform settings from the feature tokens into a tracking token, a depth token, and a visibility token in accordance with an input prompt, and logic configured to transform the tracking token, depth token, and visibility token into track predictions for an object specified by the input prompt.
Apparatuses, systems, and techniques to identify a clock frequency at which one or more processors are to operate. In at least one embodiment, a processor performs an application programming interface (API) to cause one or more one or more activity levels of one or more processors to be measured at one or more indicated intervals.
Apparatuses, systems, and techniques to select one or more discontinuous wireless communication patterns. In at least one embodiment, a processor includes one or more circuits to cause one or more different discontinuous wireless communication patterns to be selected based, at least in part, on one or more different beams with which to communicate the one or more different discontinuous wireless communication patterns.
H04W 72/231 - Canaux de commande ou signalisation pour la gestion des ressources dans le sens descendant de la liaison sans fil, c.-à-d. en direction du terminal les données de commande provenant des couches au-dessus de la couche physique, p. ex. signalisation RRC ou MAC-CE
50.
Systems and methods for remote client access to server-based software development
Systems and methods to support remote client access to server-based software development within a server that manages a container cluster, are disclosed. Exemplary implementations may launch one or more pods that may include sets of containers, including a first pod that executes a container management software application; launch a first set of containers including a first container; receive a connection instruction to establish a secure channel between a client computing platform and a remotely-accessible server-based software development environment (SDE) in the first container; establish the secure channel; receive user input for particular execution in the remotely-accessible server-based software development environment; perform the particular execution; and/or other actions.
Systems and methods include a fan and a fan controller configured to set a default speed of the fan. An initialization controller is configured to determine that the fan is powered on and to issue a control signal to the fan controller, responsive to the determination that the fan is powered on. The control signal overrides the default speed of the fan.
Various embodiments can perform efficient, large scale fluid simulation using a method, such as the fluid-implicit particle (FLIP) method, over a sparse hierarchy of grids. The grids may be represented in a number of different formats, and may include GVDB Voxels. Such approaches can handle tens of millions of particles within a virtually unbounded simulation domain. Embodiments can utilize a parallel, sparse grid hierarchy construction and provide for fast incremental updates on graphics processing unit (GPU) hardware, for example, for moving particles. In addition, a FLIP-based technique can be used to perform sparse, work-efficient parallel data gathering from particle to voxel. Various embodiments can also utilize a matrix-free GPU-based conjugate gradient solver optimized for sparse grids. Such approaches can provide orders of magnitude faster simulations on GPU hardware with respect to conventional simulations.
09 - Appareils et instruments scientifiques et électriques
42 - Services scientifiques, technologiques et industriels, recherche et conception
Produits et services
Downloadable software libraries being computer software
development tools; downloadable software and software
libraries for data compression and decompression;
downloadable software and software libraries for data
compression and decompression using graphics processing
units (GPUs); downloadable software development tools for
data compression and decompression using graphics processing
units (GPUs). Providing temporary use of online non-downloadable software
libraries being computer software development tools;
providing temporary use of online non-downloadable software
and software libraries for data compression and
decompression; providing temporary use of online
non-downloadable software and software libraries for data
compression and decompression using graphics processing
units (GPUs); providing temporary use of online
non-downloadable software development tools for data
compression and decompression using graphics processing
units (GPUs); design and development of computer software.
54.
SECURE VIRTUALIZED CRYPTOGRAPHIC SUBSYSTEMS FOR AUTONOMOUS SYSTEMS AND APPLICATIONS
In various examples, the disclosed techniques include receiving, from an application executing in a virtual machine (VM), a request to perform a cryptographic operation, wherein the request specifies an ephemeral key identifier and source data. The techniques also determine, using key metadata received from a trusted execution environment, a key slot identifier associated with the ephemeral key identifier, wherein the key slot identifier identifies a key slot in which a cryptographic key is stored. The techniques further cause the cryptographic operation to be performed on the source data in the trusted execution environment using the cryptographic key, where the cryptographic key used to perform the cryptographic operation is accessed from the key slot identified by the key slot identifier. The techniques further provide, to the application, a cryptographic operation result received form the trusted execution environment.
G06F 9/455 - ÉmulationInterprétationSimulation de logiciel, p. ex. virtualisation ou émulation des moteurs d’exécution d’applications ou de systèmes d’exploitation
Apparatuses, systems, and techniques to identify one or more objects in one or more images. In at least one embodiment, one or more objects are identified in one or more images based, at least in part, on a likelihood that one or more objects is different from other objects in one or more images.
Embodiments of the present disclosure relate to an AI agentic system for scene understanding. Some embodiments perform such scene understanding by extracting, indexing, and iteratively refining scene data through an AI agent that autonomously generates and refines queries in a continuous loop until a predefined completeness threshold is met. This ensures that scene data is not only captured but also refined over time, producing a fully indexed and queryable representation of the scene.
G06F 16/71 - IndexationStructures de données à cet effetStructures de stockage
G06F 16/787 - Recherche de données caractérisée par l’utilisation de métadonnées, p. ex. de métadonnées ne provenant pas du contenu ou de métadonnées générées manuellement utilisant des informations géographiques ou spatiales, p. ex. la localisation
57.
PIXEL BLENDING FOR NEURAL NETWORK-BASED IMAGE GENERATION
Apparatuses, systems, and techniques are presented to generate one or more images. In at least one embodiment, two or more pixels from two or more images are blended based, at least in part, on a distance of the two or more pixels from a region of the two or more images, in which pixel colors are substantially similar.
Apparatuses, systems, and techniques to optimize processor performance. In at least one embodiment, a method optimizes linked code based, at least in part, on storing an indication of whether two portions of code have been linked.
In various examples, properties may be determined for image regions, where the image regions are indicated by output data generated using MLMs. An encoder may use the properties to generate encoded images using encoding quality settings for the image regions. When an encoded image is decoded and applied to the MLMs, corresponding output data may indicate an image region which is likely to correspond to an encoded image region of the encoded image, and which may be applied to at least one MLM. Thus, the properties for encoding an image region to an encoded image can be adapted to control the visual quality of an image region determined from a decoded version of the encoded image. The properties may be determined based at least on performance metric values for the MLMs or based at least on a ranking of the image regions.
H04N 19/42 - Procédés ou dispositions pour le codage, le décodage, la compression ou la décompression de signaux vidéo numériques caractérisés par les détails de mise en œuvre ou le matériel spécialement adapté à la compression ou à la décompression vidéo, p. ex. la mise en œuvre de logiciels spécialisés
H04N 19/154 - Qualité visuelle après décodage mesurée ou estimée de façon subjective, p. ex. mesure de la distorsion
H04N 19/167 - Position dans une image vidéo, p. ex. région d'intérêt [ROI]
H04N 19/176 - Procédés ou dispositions pour le codage, le décodage, la compression ou la décompression de signaux vidéo numériques utilisant le codage adaptatif caractérisés par l’unité de codage, c.-à-d. la partie structurelle ou sémantique du signal vidéo étant l’objet ou le sujet du codage adaptatif l’unité étant une zone de l'image, p. ex. un objet la zone étant un bloc, p. ex. un macrobloc
In various examples, techniques for determining perception zones for object detection are described. For instance, a system may use a dynamic model associated with an ego-machine, a dynamic model associated with an object, and one or more possible interactions between the ego-machine and the object to determine a perception zone. The system may then perform one or more processes using the perception zone. For instance, if the system is validating a perception system of the ego-machine, the system may determine whether a detection error associated with the object is a safety-critical error based on whether the object is located within the perception zone. Additionally, if the system is executing within the ego-machine, the system may determine whether the object is a safety-critical object based on whether the object is located within the perception zone.
G05D 1/00 - Commande de la position, du cap, de l'altitude ou de l'attitude des véhicules terrestres, aquatiques, aériens ou spatiaux, p. ex. utilisant des pilotes automatiques
G06V 20/58 - Reconnaissance d’objets en mouvement ou d’obstacles, p. ex. véhicules ou piétonsReconnaissance des objets de la circulation, p. ex. signalisation routière, feux de signalisation ou routes
61.
DATA ASSOCIATION USING CORRELATION RESPONSE VALUES
In various examples, image areas may be extracted from a batch of one or more images and may be scaled, in batch, to one or more template sizes. Where the image areas include search regions used for localization of objects, the scaled search regions may be loaded into Graphics Processing Unit (GPU) memory and processed in parallel for localization. Similarly, where image areas are used for filter updates, the scaled image areas may be loaded into GPU memory and processed in parallel for filter updates. The image areas may be batched from any number of images and/or from any number of single-and/or multi-object trackers. Further aspects of the disclosure provide approaches for associating locations using correlation response values, for learning correlation filters in object tracking based at least on focused windowing, and for learning correlation filters in object tracking based at least on occlusion maps.
G06T 1/20 - Architectures de processeursConfiguration de processeurs p. ex. configuration en pipeline
G06T 11/20 - Traçage à partir d'éléments de base, p. ex. de lignes ou de cercles
G06V 10/764 - Dispositions pour la reconnaissance ou la compréhension d’images ou de vidéos utilisant la reconnaissance de formes ou l’apprentissage automatique utilisant la classification, p. ex. des objets vidéo
G06V 10/82 - Dispositions pour la reconnaissance ou la compréhension d’images ou de vidéos utilisant la reconnaissance de formes ou l’apprentissage automatique utilisant les réseaux neuronaux
G06V 20/58 - Reconnaissance d’objets en mouvement ou d’obstacles, p. ex. véhicules ou piétonsReconnaissance des objets de la circulation, p. ex. signalisation routière, feux de signalisation ou routes
In various examples, an obstacle detector is capable of tracking a velocity state of detected objects or obstacles using LiDAR data. For example, using LiDAR data alone, an iterative closest point (ICP) algorithm may be used to determine a current state of detected objects for a current frame and a Kalman filter may be used to maintain a tracked state of the one or more objects detected over time. The obstacle detector may be configured to estimate velocity for one or more detected objects, compare the estimated velocity to one or more previous tracked states for previously detected objects, determine that the detected objects corresponds to a certain previously detected object, and update the tracked state for the previously detected object with the estimated velocity.
B60T 7/12 - Organes d'attaque de la mise en action des freins par déclenchement automatiqueOrganes d'attaque de la mise en action des freins par déclenchement non soumis à la volonté du conducteur ou du passager
B60W 30/09 - Entreprenant une action automatiquement pour éviter la collision, p. ex. en freinant ou tournant
B62D 15/02 - Indicateurs de direction ou aides de direction
G01S 17/58 - Systèmes de détermination de la vitesse ou de la trajectoireSystèmes de détermination du sens d'un mouvement
G01S 17/931 - Systèmes lidar, spécialement adaptés pour des applications spécifiques pour prévenir les collisions de véhicules terrestres
63.
COMPENSATING COUPLING TRANSFORMERS FOR TRANS-INDUCTOR VOLTAGE REGULATORS
Various embodiments disclose a trans-inductor voltage regulator comprising a first group pairing of switching circuits that includes, a first group of switching circuits, a first compensating coupling inductor coupled to the first group of switching circuits, a second group of switching circuits, and a second compensating coupling inductor coupled to the second group of switching circuits, where the first compensating coupling inductor comprises a first winding of a compensating coupling transformer, and the second compensating coupling inductor is a second winding of the compensating coupling transformer.
H02M 3/335 - Transformation d'une puissance d'entrée en courant continu en une puissance de sortie en courant continu avec transformation intermédiaire en courant alternatif par convertisseurs statiques utilisant des tubes à décharge avec électrode de commande ou des dispositifs à semi-conducteurs avec électrodes de commande pour produire le courant alternatif intermédiaire utilisant des dispositifs du type triode ou transistor exigeant l'application continue d'un signal de commande utilisant uniquement des dispositifs à semi-conducteurs
64.
DISTRIBUTED SIMULATION OF QUANTUM SYSTEMS WITH OPTIMIZED TENSOR MODE REDISTRIBUTION
In various embodiments, systems and methods for distributed simulation of time dynamics of quantum systems with an optimized tensor mode redistribution schedule are provided. A scheduling optimizer may create a schedule of tensor redistributions based on a hypergraph constructed from a quantum many-body operator. The scheduling optimizer iteratively distributes a quantum state tensor into state tensor slices based on a clustering process that partitions the hypergraph into clusters, and redistributes the quantum state tensor across processing units based on identified sliced or non-sliced tensor modes according to the obtained clusters. Operators that can be applied to non-sliced modes of the redistributed quantum state tensor are applied, and the iterative process repeats until each of the operators of the full quantum many-body operator have been applied. Because these operators act on non-sliced modes, inter-processing unit communication is minimized while executing these operations, and tensor redistributions across all processing units is minimized.
Approaches presented herein provide for the use of implicit functions to simulate differentiable object elasticity. An implicit continuous function, such as a signed distance function (SDF), can be used to approximate the surface of an object by providing scalar values from a set of vertices of a regular grid in which the object representation is to be generated. Interpolation can be applied to determine an approximate surface location and shape within each boundary cell. A trained neural network, such as a multilayer perceptron (MLP), can be used to determine appropriate quadrature points that fall within the volume of the object. A finite element analysis can integrate over these quadrature points, using both continuous and discrete settings, as a basis for performing efficient differentiable elasticity simulations including the deformable object.
Systems and methods are disclosed that monitor for supply noise from a power source using a voltage noise monitor (VNM). For instance, the VNM may include voltage sense circuitry comprising a controllable resistor that is controlled using threshold information. The resistance of the controllable resistor may be changed based on closing and/or opening one or more switches associated with step resistors using the bits from the threshold information. Furthermore, the VNM may include digital circuitry that comprises a hold finite state machine and a sticky hold counter. Using the digital circuitry, the VNM may be configured to hold a noise detection event for a plurality of clock cycles. In addition, the VNM may perform a calibration process based on setting two voltages for the power source to obtain two codes, and determining a transfer function based on the two voltages and the two codes.
Various embodiments include techniques for encoding media frames using a multistage search without the processing overhead of pyramidal motion estimation techniques. The first stage of the multistage search generates many motion vectors, where each motion vector is based on a small range full pixel search of a pixel group in the media frame. The video encoder selects a full pixel best motion (FBM) vector for all pixel groups from the motion vectors generated during the first stage. The second and subsequent phases of the multistage search is based on the FBM vector from the prior stage as a starting point, where the search range for each subsequent phase is larger than the search range of the prior stage. The multistage search can be performed over a fixed number of stages. Alternatively, the multistage search can be terminated when the cost value for the current stage is below a threshold value.
Apparatuses, systems, and techniques to translate instructions that use operand stack structures to instructions that use registers and/or memory for operands. In at least one embodiment, one or more instruction operand stack structures are duplicated if one or more branch instructions use one or more instruction operands stored in the one or more stack structures.
A system includes a gain module, an optical module, and a feedback module. The gain module processes a first portion of an electrical signal to generate a first compensated portion of the electrical signal, the first portion of the electrical signal having a first frequency range. The optical module generates an optical signal based on a combination of the first compensated portion of the electrical signal and a second portion of the electrical signal having a second frequency range that is higher than the first frequency range. The feedback module provides, to the gain module, an electrical feedback signal based at least in part on a signal strength of the optical signal. The gain module updates the first compensated portion of the electrical signal based on the electrical feedback signal.
Apparatuses, systems, and techniques to perform one or more operations using a tensor. In at least one embodiment, one or more circuits are to perform an application programming interface (API) to cause two or more tensor contractions to be performed based, at least in part, on one or more input parameters of the API.
Apparatuses, systems, and techniques to cause one or more non-unique wireless device identifiers to correspond to one or more unique wireless device identifiers. In at least one embodiment, one or more non-unique wireless device identifiers are mapped to one or more unique wireless device identifiers.
Various embodiments include a printed circuit board cover assembly comprising a cover layer that attaches to at least a portion of a printed circuit board layer, and a retention hook that extends from the cover layer and physically interfaces with at least a portion of a connection port. In various embodiments, a printed circuit board package comprises a printed circuit board layer, a cover layer that attaches to at least a portion of the printed circuit board layer, and a retention hook extending from the cover layer, the retention hook configured to physically interface with at least a portion of a connection port.
Apparatuses, systems, and methods to store non-graphics kernels with graphics kernels. In at least one embodiment, performance of an application programming interface (“API”) causes non-graphics kernels to be stored sequentially with graphics kernels to be performed by one or more processors.
Apparatuses, systems, and techniques to perform a digital simulation of a wireless network. In at least one embodiment, a processor generates a virtual network simulation and combine user equipment of said virtual network simulation into virtual cells in order to cause at least one of two or more mobile device emulation programs to be selected based, at least in part, on two or more distinct physical resource block parameters corresponding to said two or more mobile devices.
Disclosed are apparatuses, systems, and techniques that implement training and deployment of streaming multimodal language systems capable of generating live text outputs. The techniques include predicting, over a plurality of iterations, a plurality of text tokens of a streaming text output associated with a streaming audio input. An individual iteration updates audio embeddings representative of the streaming audio input, processes, using a cross-modality network, the audio embeddings and text embeddings representative of a text input associated with the streaming audio input to obtain a plurality of cross-attention states, provides, to a language model (LM), a prompt including output embeddings obtained based at least on the plurality of cross-attention states, and receives, from the LM, a text token predicted for the respective iteration. The streaming text output is generate using the predicted text tokens.
G10L 15/183 - Classement ou recherche de la parole utilisant une modélisation du langage naturel selon les contextes, p. ex. modèles de langage
G10L 15/06 - Création de gabarits de référenceEntraînement des systèmes de reconnaissance de la parole, p. ex. adaptation aux caractéristiques de la voix du locuteur
G10L 15/30 - Reconnaissance distribuée, p. ex. dans les systèmes client-serveur, pour les applications en téléphonie mobile ou réseaux
Apparatuses, systems, and methods to perform matrix multiplication. In at least one embodiment, a matrix multiplication is performed based on an indication of whether information to be used by part of the matrix multiplication has been loaded.
Apparatuses, systems, and methods to perform matrix multiplication. In at least one embodiment, a processor performs an API to indicate whether a partial result of a matrix multiplication has been loaded.
Apparatuses, systems, and methods to perform matrix multiplication. In at least one embodiment, a processor performs an API to indicate a portion of results of a matrix multiply operation that is to be partially loaded.
Apparatuses, systems, and methods to perform matrix multiplication. In at least one embodiment, a processor performs an API to indicate operands of a matrix multiply operation that are to be partially loaded.
Apparatuses, systems, and methods to perform matrix multiplication. In at least one embodiment, a processor performs an API to indicate whether operands to be used by a matrix multiplication have been partially loaded.
Apparatuses, systems, and techniques to perform one or more operations using a tensor. In at least one embodiment, one or more circuits are to perform an application programming interface (API) to cause parameters indicating operands and dimensions of operands of two or more tensor contractions to be stored.
Apparatuses, systems, and techniques to help identify one or more directions to transmit a first fifth generation new radio (“5G NR”) signal. In at least one embodiment, said one or more identified directions to be used to transmit a first 5G NR signal is based, at least in part, on channel state information of one or more second 5G NR signals.
H04B 7/06 - Systèmes de diversitéSystèmes à plusieurs antennes, c.-à-d. émission ou réception utilisant plusieurs antennes utilisant plusieurs antennes indépendantes espacées à la station d'émission
83.
INSTRUCTION GENERATION USING ONE OR MORE NEURAL NETWORKS
Apparatuses, systems, and techniques are presented for generating instructional text. In at least one embodiment, an instructional video is analyzed to determine logical steps of a process or task demonstrated in that video, and instructive text is generated for those logical steps.
G06V 10/82 - Dispositions pour la reconnaissance ou la compréhension d’images ou de vidéos utilisant la reconnaissance de formes ou l’apprentissage automatique utilisant les réseaux neuronaux
G06V 10/94 - Architectures logicielles ou matérielles spécialement adaptées à la compréhension d’images ou de vidéos
G06V 20/40 - ScènesÉléments spécifiques à la scène dans le contenu vidéo
G06V 30/19 - Reconnaissance utilisant des moyens électroniques
G06V 30/262 - Techniques de post-traitement, p. ex. correction des résultats de la reconnaissance utilisant l’analyse contextuelle, p. ex. le contexte lexical, syntaxique ou sémantique
G09B 9/00 - Simulateurs pour l'enseignement ou l'entraînement
84.
LANGUAGE MODEL PERFORMANCE ON LOW RESOURCE LANGUAGES
In various examples, techniques are described for adapting a multilingual Large Language Model (LLM) into a bilingual Small Language Model (SLM) that exhibits model capacity to understand, process, and generate content in both English and a Low-Resource Language (LRL). The techniques include compressing the LLM to generate a multilingual SLM and performing continued pre-training on the multilingual SLM to generate the bilingual SLM. The techniques also include performing one or more alignment techniques on the bilingual SLM to adapt the SLM's outputs to human values and expectations regarding, e.g., profanity, privacy, politeness, bias, and/or conversational style. The techniques may generate various training corpora, each including one or more of natural English content, natural LRL content, synthetic LRL content generated via translation from English sources, and transliterated synthetic LRL content based on transliterations of natural and/or synthetic LRL content.
G06F 40/58 - Utilisation de traduction automatisée, p. ex. pour recherches multilingues, pour fournir aux dispositifs clients une traduction effectuée par le serveur ou pour la traduction en temps réel
85.
LONG-RANGE 3D OBJECT DETECTION USING 2D BOUNDING BOXES
3D object detection is a computer vision task that generally detects (e.g. classifies and localizes) objects in 3D space from the 2D images or videos that capture the objects. Current techniques used for 3D object detection rely on machine learning processes that learn to detect 3D objects from existing images annotated with high-quality 3D information including depth information generally obtained using lidar technology. However, due to lidar's limited measurable range, current machine learning solutions to 3D object detection do not support detection of 3D objects beyond the lidar range, which is needed for numerous applications, including autonomous driving applications where existing close or midrange 3D object detection does not always meet the safety-critical requirement of autonomous driving. The present disclosure provides for 3D object detection using a technique that supports long-range detection (i.e. detection beyond the lidar range).
G06T 7/50 - Récupération de la profondeur ou de la forme
G06T 7/70 - Détermination de la position ou de l'orientation des objets ou des caméras
G06T 7/80 - Analyse des images capturées pour déterminer les paramètres de caméra intrinsèques ou extrinsèques, c.-à-d. étalonnage de caméra
G06V 10/22 - Prétraitement de l’image par la sélection d’une région spécifique contenant ou référençant une formeLocalisation ou traitement de régions spécifiques visant à guider la détection ou la reconnaissance
G06V 10/82 - Dispositions pour la reconnaissance ou la compréhension d’images ou de vidéos utilisant la reconnaissance de formes ou l’apprentissage automatique utilisant les réseaux neuronaux
In various examples, techniques for optimizing object detection models are described herein. Systems and methods are disclosed that process sensor data using a backbone of a machine learning model(s) in order to generate feature maps at different resolutions. The systems and methods then use the machine learning model(s) to generate a vector based at least in part on one or more of the feature maps. For example, if the backbone generates four feature maps, then the machine learning model(s) may generate the vector using two feature maps from the four feature maps. The systems and methods then process the vector using a transformer of the machine learning model(s) in order to generate data representing a class label(s) for an object(s) depicted by an image represented by the sensor data and/or a location(s) of the object(s) within the image.
G06V 10/82 - Dispositions pour la reconnaissance ou la compréhension d’images ou de vidéos utilisant la reconnaissance de formes ou l’apprentissage automatique utilisant les réseaux neuronaux
G06V 10/77 - Traitement des caractéristiques d’images ou de vidéos dans les espaces de caractéristiquesDispositions pour la reconnaissance ou la compréhension d’images ou de vidéos utilisant la reconnaissance de formes ou l’apprentissage automatique utilisant l’intégration et la réduction de données, p. ex. analyse en composantes principales [PCA] ou analyse en composantes indépendantes [ ICA] ou cartes auto-organisatrices [SOM]Séparation aveugle de source
Apparatuses, systems, and techniques to generate an image. In at least one embodiment, one or more neural networks are to generate a second image based, at least in part, on a first image and information indicating zero or more differences between the first and second image.
G06V 20/40 - ScènesÉléments spécifiques à la scène dans le contenu vidéo
G06V 10/82 - Dispositions pour la reconnaissance ou la compréhension d’images ou de vidéos utilisant la reconnaissance de formes ou l’apprentissage automatique utilisant les réseaux neuronaux
Technologies for generating a set of models for each account, where each model is a fine-grained, unsupervised behavior model trained for each user to monitor and detect anomalous patterns are described. An unsupervised training pipeline can generate user models, each being associated with one of multiple accounts and is trained to detect an anomalous pattern using feature data associated with the one account. Each account is associated with at least one of a user, a machine, or a service. An inference pipeline can detect a first anomalous pattern in first data associated with a first account using a first user model. The inference pipeline can detect a second anomalous pattern in second data associated with a second account using a second user model.
A method includes providing heat produced by a fuel cell from converting a source of gas to generate electrical power for a datacenter to an absorption chiller to produced cooled liquid. The method further includes utilizing the produced cooled liquid to cool one or more electronic components of the datacenter.
H05K 7/20 - Modifications en vue de faciliter la réfrigération, l'aération ou le chauffage
G06F 1/26 - Alimentation en énergie électrique, p. ex. régulation à cet effet
H01M 8/04007 - Dispositions auxiliaires, p. ex. pour la commande de la pression ou pour la circulation des fluides relatives à l’échange de chaleur
H01M 8/04089 - Dispositions pour la commande des paramètres des réactifs, p. ex. de la pression ou de la concentration des réactifs gazeux
H01M 8/04119 - Dispositions pour la commande des paramètres des réactifs, p. ex. de la pression ou de la concentration des réactifs gazeux avec apport simultané ou évacuation simultanée d’électrolyteHumidification ou déshumidification
H05K 7/14 - Montage de la structure de support dans l'enveloppe, sur cadre ou sur bâti
90.
NEURAL NETWORK TRAINING USING RESAMPLED IMAGE DATA
Apparatuses, systems, and techniques to modify a set of training data used for machine learning. In at least one embodiment, a set of images used for training a machine learning system is resampled by augmenting the set of images with additional images of under represented object types extracted from portions of existing training images in the set.
G06F 18/2411 - Techniques de classification relatives au modèle de classification, p. ex. approches paramétriques ou non paramétriques basées sur la proximité d’une surface de décision, p. ex. machines à vecteurs de support
G06F 18/214 - Génération de motifs d'entraînementProcédés de Bootstrapping, p. ex. ”bagging” ou ”boosting”
Apparatuses, systems, and techniques to perform one or more operations using a tensor. In at least one embodiment, one or more circuits are to perform an application programming interface (API) to cause parameters indicating operands and dimensions of operands of two or more tensor contractions to be stored. In at least one embodiment, one or more circuits are to perform an application programming interface (API) to cause two or more tensor contractions to be performed based, at least in part, on one or more input parameters of the API.
Apparatuses, systems, and methods to perform matrix multiplication. In at least one embodiment, a matrix multiplication is performed based on an indication of whether information to be used by part of the matrix multiplication has been loaded.
Systems and methods for a datacenter cooling system are disclosed. In at least one embodiment, a secondary cooling loop interfaces with a primary cooling loop and includes at least one processor that is adapted with a first proportional-integral-derivative (PID) controller to control a first flow controller, which can cause a flow rate for a secondary coolant from a coolant distribution unit (CDU) to a plurality of second flow controllers, and where the plurality of second flow controllers are associated with a plurality of second PID controllers and the flow rate is based in part of on feedback from the plurality of second PID controllers.
Aspects of the present invention are directed to computer-implemented techniques for improving the training of artificial neural networks using a reduced precision (e.g., float16) data format. Embodiments of the present invention rescale tensor values prior to performing matrix operations (such as matrix multiplication or matrix addition) to prevent overflow and underflow. To preserve accuracy throughout the performance of the matrix operations, the scale factors are defined using a novel data format to represent tensors, wherein a matrix is represented by the tuple X, where X=(a, v[.]), wherein a is a float scale factor and v[.] are scaled values stored in the float16 format. The value of any element X[i] according to this data format would be equal to a*v[i].
In various examples, systems and methods are disclosed relating to dexterous arm-hand grasping with geometric fabrics. One or more processors can update, during a first update stage, a teacher model to generate first actions for a geometric fabric associated with a simulated autonomous machine of a simulation using state information of the simulation. During a second update stage, the processors can update a student model to generate second actions for the geometric fabric using at least one rendered image of the simulation, the teacher model, and noised state information of the simulation. The student model and the geometric fabric can be used to control a physical autonomous machine with respect to a physical object based at least on an image of an environment including the physical autonomous machine and the physical object.
In various examples, systems and methods are disclosed relating to disclosed relating to dexterous arm-hand grasping with geometric fabrics. One or more processors can cause a teacher model to generate first actions for a geometric fabric associated with a simulated autonomous machine in a simulated environment using state information of the simulated environment and position information of a simulated object in the simulated environment. Using the teacher model and a depth image of the simulated environment, a student model can be updated to generate second actions for the geometric fabric associated with the simulated autonomous machine. A depth image of an environment can be provided as input to the student model to cause the student model to infer at least one action to control a physical autonomous machine with respect to a physical object using the geometric fabric.
The hybrid-head architecture model can be used to train a language model (LM). It uses a combination of attention heads and state space models (SSMs) to improve the speed and efficiency of inferencing a received input sequence. This disclosure combines the high-resolution recall capabilities of attention heads with the efficient context summarization of SSM heads. The model can be separated into a set of layers, and the input sequence can be processed layer by layer. Each layer can have its own number of attention heads and SSM heads. Fine-tuning and optimization can be applied to each layer, as well as normalization and scaling. To further optimize the performance of the hybrid-head architecture model, learnable meta tokens can be used, which act as a learned cache for attention and SSM heads, enhancing the model's focus on salient information. The attention heads and the SSMs can be processed in parallel.
Apparatuses, systems, and techniques to generate a memory allocation plan for a set of tensors. In at least one embodiment, tensor data corresponding to the set of tensors is stored into memory locations at run time based, at least in part, on the memory allocation plan generated at a compile time.
Systems and methods are disclosed for training and using a digital human foundational model (DHFM) comprising a generative adversarial network (GAN) generator. For instance, the method may include obtaining one or more inputs comprising pose information indicating a three-dimensional (3D) pose representation of a human and processing the one or more inputs using a mapping network to generate intermediate latent code. The method may further include processing the intermediate latent code using the trained generator to generate texel-aligned Gaussian maps that align Gaussian attributes to a coarse mesh template of the human and performing linear blend skinning and deformation on the texel-aligned Gaussian maps to obtain modified texel-aligned Gaussian maps. The method may also include processing the modified texel-aligned Gaussian maps using a multi-part renderer to generate a synthetic human representation of the human indicating facial and hand features of the human.
G06T 7/246 - Analyse du mouvement utilisant des procédés basés sur les caractéristiques, p. ex. le suivi des coins ou des segments
G06T 13/40 - Animation tridimensionnelle [3D] de personnages, p. ex. d’êtres humains, d’animaux ou d’êtres virtuels
G06V 10/764 - Dispositions pour la reconnaissance ou la compréhension d’images ou de vidéos utilisant la reconnaissance de formes ou l’apprentissage automatique utilisant la classification, p. ex. des objets vidéo
G06V 10/774 - Génération d'ensembles de motifs de formationTraitement des caractéristiques d’images ou de vidéos dans les espaces de caractéristiquesDispositions pour la reconnaissance ou la compréhension d’images ou de vidéos utilisant la reconnaissance de formes ou l’apprentissage automatique utilisant l’intégration et la réduction de données, p. ex. analyse en composantes principales [PCA] ou analyse en composantes indépendantes [ ICA] ou cartes auto-organisatrices [SOM]Séparation aveugle de source méthodes de Bootstrap, p. ex. "bagging” ou “boosting”
G06V 10/776 - ValidationÉvaluation des performances
G06V 10/82 - Dispositions pour la reconnaissance ou la compréhension d’images ou de vidéos utilisant la reconnaissance de formes ou l’apprentissage automatique utilisant les réseaux neuronaux
G06V 40/16 - Visages humains, p. ex. parties du visage, croquis ou expressions
100.
GENERATIVE THREE-DIMENSIONAL (3D) DIGITAL HUMAN FOUNDATION MODEL FROM IN THE WILD TWO-DIMENSIONAL (2D) IMAGES
Systems and methods are disclosed for generating and curating a training dataset for training one or more machine learning-artificial intelligence (ML-AI) models. For instance, the method may include extracting 2D landmarks of a human from an obtained image that is within the training dataset and extracting 3D poses of the human from the obtained image. The method may further include using camera coordinates associated with the obtained image to project the 3D poses of the human into 2D space and fine-tuning the 3D poses of the human based on comparing the projected 3D poses in 2D space with the extracted 2D landmarks. The method may also include generating labels for the obtained image within the training dataset, augmenting the training dataset with a plurality of generated synthetic images of humans, and training the one or more ML-AI models.
G06T 7/246 - Analyse du mouvement utilisant des procédés basés sur les caractéristiques, p. ex. le suivi des coins ou des segments
G06T 13/40 - Animation tridimensionnelle [3D] de personnages, p. ex. d’êtres humains, d’animaux ou d’êtres virtuels
G06V 10/26 - Segmentation de formes dans le champ d’imageDécoupage ou fusion d’éléments d’image visant à établir la région de motif, p. ex. techniques de regroupementDétection d’occlusion
G06V 10/72 - Préparation de données, p. ex. prétraitement statistique des caractéristiques d’images ou de vidéos
G06V 10/764 - Dispositions pour la reconnaissance ou la compréhension d’images ou de vidéos utilisant la reconnaissance de formes ou l’apprentissage automatique utilisant la classification, p. ex. des objets vidéo
G06V 10/774 - Génération d'ensembles de motifs de formationTraitement des caractéristiques d’images ou de vidéos dans les espaces de caractéristiquesDispositions pour la reconnaissance ou la compréhension d’images ou de vidéos utilisant la reconnaissance de formes ou l’apprentissage automatique utilisant l’intégration et la réduction de données, p. ex. analyse en composantes principales [PCA] ou analyse en composantes indépendantes [ ICA] ou cartes auto-organisatrices [SOM]Séparation aveugle de source méthodes de Bootstrap, p. ex. "bagging” ou “boosting”
G06V 10/776 - ValidationÉvaluation des performances
G06V 10/94 - Architectures logicielles ou matérielles spécialement adaptées à la compréhension d’images ou de vidéos
G06V 20/70 - Étiquetage du contenu de scène, p. ex. en tirant des représentations syntaxiques ou sémantiques
G06V 40/16 - Visages humains, p. ex. parties du visage, croquis ou expressions
