Implementations relate to an automated assistant that can determine whether to respond to inputs in an environment according to whether radar data indicates a user is present. When user presence is detected, the automated assistant can virtually segment the environment and apply certain operational parameters to certain segments of the environment. For instance, the automated assistant can enable an input detection feature, such as warm word detection, for a segmented portion of the environment in which a user is detected. In this way, false positives can be mitigated for instances in which environmental and/or user sounds are detected by the automated assistant but do not originate from a particular segment of the environment. Other parameters, such as varying confidence thresholds and/or speech processing biasing, can be temporarily enforced for different segments of an environment in which a user is detected.
A UE (102) receives (305), from a network entity (106a), a configuration for a TCI state list for a first serving cell that corresponds to a reference cell. The UE (102) further receives (309), from the network entity (106a), a first parameter and a second parameter that define an other TCI state list for a second serving cell. The first parameter indicates a type of the other TCI state list for the second serving cell and the second parameter indicates at least one of: a serving cell index for the first serving cell or one or more TCI state IDs in the TCI state list. The UE (102) communicates (377) with the network entity (106a) using the other TCI state list for the second serving cell. The other TCI state list is based on the second parameter and the TCI state list for the first serving cell.
A computing device may drive a haptic device of the computing device to output a precursor haptic signal. The computing device may determine a motion signal associated with outputting the precursor haptic signal, lire computing device may determine, based at least in part on the motion signal associated with outputting the precursor haptic signal, that the computing device is in an adverse haptic environment. The computing device may, in response to determining that the computing device is in an adverse haptic environment, drive, by the one or more processors, the haptic device to output an alternative haptic signal instead of the haptic signal.
An enhanced matrix product state-based decoder is generated and employed to almost optimally detect and correct errors within a quantum computing and information processing system. The decoder takes as input a detector level error model that describes physical error channels and a set of error detections. This error model is improved using experimental data.
5.
DUAL BAND WIRELESS COMMUNICATIONS FOR MULTIPLE CONCURRENT AUDIO STREAMS
Various arrangements for performing wireless device-to-device communication are presented. An audio output device, such as an earbud or pair of earbuds, can establish a connection with an audio source via a first Bluetooth interface that communicates using a Bluetooth communication protocol on a 2.4 GHz Bluetooth frequency band. The audio output device can negotiate that Bluetooth frequency-shifted communication, such as on a 5 or 6 GHz frequency band, is available for use with the audio source. The audio output device may then perform Bluetooth frequency-shifted communication with the audio source such that the audio output device receives an audio stream from the audio source using Bluetooth frequency-shifted communication and the Bluetooth communication protocol.
H04W 4/80 - Services using short range communication, e.g. near-field communication [NFC], radio-frequency identification [RFID] or low energy communication
Filter coefficient derivation simplification for cross-component prediction reduces latencies typically introduced by convolutional cross-component model (CCCM) prediction and thus enables use of CCCM prediction by hardware coders. Various approaches for filter coefficient derivation simplification are disclosed, including limiting a dynamic range of filter coefficient derivation to a defined bit range, limiting filter coefficient derivation and thus use of CCCM prediction based on coding unit size, and/or enabling filter coefficient derivation directly from non-downsampled luma samples.
H04N 19/105 - Selection of the reference unit for prediction within a chosen coding or prediction mode, e.g. adaptive choice of position and number of pixels used for prediction
H04N 19/157 - Assigned coding mode, i.e. the coding mode being predefined or preselected to be further used for selection of another element or parameter
H04N 19/176 - Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object the region being a block, e.g. a macroblock
H04N 19/186 - Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being a colour or a chrominance component
H04N 19/593 - Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding involving spatial prediction techniques
H04N 19/70 - Methods or arrangements for coding, decoding, compressing or decompressing digital video signals characterised by syntax aspects related to video coding, e.g. related to compression standards
Region-based cross-component prediction improves convolutional cross-component mode (CCCM) prediction by enabling filter coefficients for predicting chroma samples from luma samples to be derived for an entire region of a frame of a video stream, such as a coding tree unit (CTU), rather than requiring that such filter coefficients be derived for each individual coding unit (CU). Deriving the filter coefficients for an entire region instead of for each individual CU under processing significantly reduces the latency in video coding and thus enables CCCM prediction to be used in hardware coder implementations.
H04N 19/105 - Selection of the reference unit for prediction within a chosen coding or prediction mode, e.g. adaptive choice of position and number of pixels used for prediction
H04N 19/157 - Assigned coding mode, i.e. the coding mode being predefined or preselected to be further used for selection of another element or parameter
H04N 19/176 - Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object the region being a block, e.g. a macroblock
H04N 19/186 - Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being a colour or a chrominance component
H04N 19/593 - Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding involving spatial prediction techniques
H04N 19/70 - Methods or arrangements for coding, decoding, compressing or decompressing digital video signals characterised by syntax aspects related to video coding, e.g. related to compression standards
8.
TRANSLATION AND SCALING EQUIVARIANT SLOT ATTENTION
A method includes receiving feature vectors and, for each respective feature vector, a corresponding absolute positional encoding. The method also includes determining latent representations of entities represented by the feature vectors, and determining, for each respective latent representation, a corresponding relative positional encoding based on the corresponding absolute positional encoding of each feature vector and a corresponding position vector associated with the respective latent representation. The method additionally includes determining an attention matrix based on the feature vectors, the entity-centric latent representations, and the corresponding relative positional encoding of each latent representation. The method further includes updating, for each respective latent representation, the corresponding position vector based on a weighted mean of the corresponding absolute positional encoding of each feature vector weighted according to corresponding entries of the attention matrix, and outputting the latent representations and/or the position vectors associated therewith.
G06N 3/0442 - Recurrent networks, e.g. Hopfield networks characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
A waveguide includes a plurality of reflective facet sets. Each reflective facet set of the plurality of reflective facet sets includes a first reflective facet to reflect light having a first optical characteristic and a second reflective facet to reflect light having a second optical characteristic that is different from the first optical characteristic. A first reflective facet in a first reflective facet set of the plurality of reflective facet sets overlaps a first reflective facet of a second set of the plurality of reflective facet sets.
A method (400) for an aggregatable application programming interface (API) includes receiving, from a third party service (150), an aggregation request (20) requesting aggregation of client data (30) from a client (12) of the third party service. The method also includes receiving, from an API (14) executed by a client device (10) of the client, a first portion of the client data (30a). The method includes storing the first portion of the client data and receiving, from the API, a second portion of the client data (30b). The method includes determining that the second portion of the client data is a final portion of the client data. In response, the method includes aggregating the first portion of the client data with the second portion of the client data. The method also includes transmitting the aggregated client data (30A) to the third party service.
Decoding a current block includes receiving a compressed bitstream. A transform block of transform coefficients is decoded from the compressed bitstream. The transform coefficients are in a transform domain. The transform block is input to a machine-learning model to obtain a residual block that is in a pixel domain. The residual block is used to reconstruct the current block. Encoding a current block includes receiving a current residual block. The current residual block and a specified rate-distortion parameter are input to a machine-learning model to obtain a quantized transform block. The quantized transform block is entropy encoded into a compressed bitstream.
H04N 19/107 - Selection of coding mode or of prediction mode between spatial and temporal predictive coding, e.g. picture refresh
H04N 19/147 - Data rate or code amount at the encoder output according to rate distortion criteria
H04N 19/176 - Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object the region being a block, e.g. a macroblock
H04N 19/18 - Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being a set of transform coefficients
H04N 19/91 - Entropy coding, e.g. variable length coding [VLC] or arithmetic coding
12.
RENDERING AUGMENTED REALITY CONTENT BASED ON POST-PROCESSING OF APPLICATION CONTENT
Implementations relate to an automated assistant that provides augmented reality content, via a display interface of computerized glasses, resulting from post-processing of application content. The application content can be identified based on prior interactions between a user and one or more applications, and the application content can be processed to determine objects, and/or object classifications, that may be associated with the application content. When the user is wearing the computerized glasses, and the object is detected within a field of view of the computerized glasses, the automated assistant can cause certain content to be rendered at the display interface of the computerized glasses. In some implementations, the content can be generated to supplement, and/or be different from, existing content that the user may have already accessed, in furtherance of preventing duplicative usage of applications and/or preserving computational resources.
Implementations relate to an automated assistant that can proactively detect and respond to a request for credentials. Characteristics of an entity requesting the credentials can be preemptively determined by the automated assistant using data that may be provided by the user or other previous visitors to a location. For example, the automated assistant can determine that the entity may expressly request certain information from a user when the user arrives at the location. Based on this determination, the automated assistant can operate to initialize an interface of a computing device of the user, when the user is determined to be at or near the location. For example, an audio interface of the computing device can be initialized to capture an audible request from a person who views credentials before granting access to a feature of the location.
G06F 16/909 - Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using geographical or spatial information, e.g. location
14.
A GENERALIST FRAMEWORK FOR PANOPTIC SEGMENTATION OF IMAGES AND VIDEOS
Provided are systems and methods for performing panoptic segmentation of images and videos using a denoising diffusion model. The panoptic segmentation task is formulated as a conditional discrete data generation problem. This is achieved by learning a generative model for panoptic masks, for example treated as an array of discrete tokens, conditioned on an input image. The generative model can also be applied to video data by including predictions from past frames as an additional conditioning signal. This enables the model to learn to track and segment objects automatically across video frames.
Methods and techniques for manipulating the color of an image based on a text-based description are presented herein. A system can access an input image and an input text. The system can process, using a machine-learned recolorizing model, the input image to generate a recolorized image. A system can determine the similarity between the recolorized image and the input text description using a loss function and pre-trained encoder(s) which have been trained on a large dataset of text and images to convert the text and image inputs into the same embedding space. The system can then modify the one or more parameter values of the machine-learned recolorizing model to minimize the value of the loss function. Thus, after a plurality of iterations, the machine-learned recolorizing model will generate a recolorized photo that matches the description given in the input text.
Systems and method for routing data packets in ring network. A data packet being transmitted to a destination node may be received by a first structure at a first node. The first node may determine a number of hops the data packet will traverse as it is transmitted from the first node to the destination node and compare the determined number of hops to a threshold hop value to determine whether the number of hops is equal to or less than the threshold hop value. If the number of hops is greater than the threshold, the data packet may be transmitted to a dimension queuing structure for a first virtual channel within a second node, otherwise, the data packet may be transmitted to a dimension queuing structure for a second virtual channel or a turn queuing structure within the second node.
Systems and methods provided for restricting SN status changes. UE generates measurement report B1. Measurement report B1 would, notwithstanding SN status change, trigger SN addition procedure. Determination is made whether the flag within the eNB/gNB for "Restrict-secondary-node-addition" is set to true. If determination is no, then no SN status change restriction is permitted for UE and process proceeds to perform secondary node addition. If determination is yes, determination is made whether the UE has been in connected mode within this cell for time greater than the "time-threshold-for-reject-secondary node-addition" parameter value. If determination is no, perform secondary node addition. If determination is yes, determination is made whether the UE has requested secondary node addition a number of times greater than "number-threshold-for-secondary-node-rejection". If determination is no, performs secondary node addition. If determination is yes, i.e., the eNB/gNB MN determines to not proceed with the secondary node addition request for the UE-reported measurement.
Decoding a current block using inter prediction with filtering includes identifying an intermediate prediction block for the current block using a motion vector and a reference frame. Filter coefficients are obtained for a filter. The filter coefficients are obtained using reconstructed pixels and second reconstructed pixels. The reconstructed pixels are peripheral to the current block. The second reconstructed pixels are peripheral to the intermediate prediction block. The filter is applied to the intermediate prediction block to obtain a final prediction block. The current block is reconstructed using the final prediction block. Encoding a current block includes obtaining an intermediate motion vector for the current block. Filter coefficients are obtained by minimizing an error metric between a prediction block corresponding to the intermediate motion vector and the current block. A motion vector is obtained for the current block by refining the intermediate motion vector using the filter coefficients.
H04N 19/117 - Filters, e.g. for pre-processing or post-processing
H04N 19/137 - Motion inside a coding unit, e.g. average field, frame or block difference
H04N 19/176 - Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object the region being a block, e.g. a macroblock
H04N 19/186 - Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being a colour or a chrominance component
H04N 19/196 - Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the adaptation method, adaptation tool or adaptation type used for the adaptive coding being specially adapted for the computation of encoding parameters, e.g. by averaging previously computed encoding parameters
H04N 19/46 - Embedding additional information in the video signal during the compression process
H04N 19/463 - Embedding additional information in the video signal during the compression process by compressing encoding parameters before transmission
H04N 19/82 - Methods or arrangements for coding, decoding, compressing or decompressing digital video signals - Details of filtering operations specially adapted for video compression, e.g. for pixel interpolation involving filtering within a prediction loop
19.
COLOR DECORRELATION IN VIDEO AND IMAGE COMPRESSION
Image and video compression using color decorrelation is described. A method described herein includes receiving color transform information for an encoded block of image data, wherein the color transform information identifies an adaptive transform matrix used to convert an original block of the image data from an original color space to a new color space, thereby resulting in color decorrelation of the original block. A decoder receives a compressed bitstream including the encoded block that was encoded using the new color space and reconstructs the block from the encoded block. The method includes determining, from the color transform information, the adaptive transform matrix. After reconstructing the block, an inverse color transform of the block is performed using the matrix to obtain pixel values for a reconstructed block corresponding to the original block in the original color space, and the image data including the reconstructed block is stored or transmitted.
H04N 19/12 - Selection from among a plurality of transforms or standards, e.g. selection between discrete cosine transform [DCT] and sub-band transform or selection between H.263 and H.264
H04N 19/157 - Assigned coding mode, i.e. the coding mode being predefined or preselected to be further used for selection of another element or parameter
H04N 19/176 - Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object the region being a block, e.g. a macroblock
H04N 19/186 - Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being a colour or a chrominance component
H04N 19/463 - Embedding additional information in the video signal during the compression process by compressing encoding parameters before transmission
H04N 19/70 - Methods or arrangements for coding, decoding, compressing or decompressing digital video signals characterised by syntax aspects related to video coding, e.g. related to compression standards
20.
PHYSICAL LAYER IMPROVEMENTS FOR SHORT RANGE WIRELESS COMMUNICATIONS
Various arrangements are presented that provide improvements of short-range wireless communications, such as Bluetooth LE Audio communication. An audio source device may determine that unidirectional audio is to be output. In response to determining that unidirectional audio is to be output, a first physical layer (PHY) configuration can be set for a first communication link in the downlink direction from the audio source device to the audio output device. A second PHY configuration can be set for the communication link in the uplink direction from the audio output device to the audio source device. The first PHY configuration has a greater symbol rate than the second PHY configuration.
A method (500) includes, for each training sample (410) of a plurality of training samples: processing, using a sequence transduction model (200), corresponding training input features (415) to obtain one or more output token sequence hypotheses (432) each including one or more predicted common tokens (204); and determining a token-level loss (462) based on, for each hypothesis: a number of special token insertions each associated with a corresponding predicted special token that appears in the hypothesis but does not appear in a corresponding sequence of ground-truth output tokens; and a number of special token deletions each associated with a corresponding ground-truth special token in the set of ground-truth special tokens that does not appear in hypothesis. The method also includes training the sequence transduction model to minimize additive error rate based on the token-level losses determined for the plurality of training samples.
G10L 17/04 - Training, enrolment or model building
G10L 15/06 - Creation of reference templates; Training of speech recognition systems, e.g. adaptation to the characteristics of the speaker's voice
G06N 3/0442 - Recurrent networks, e.g. Hopfield networks characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
Methods, systems, and apparatus, including computer programs encoded on computer storage media, for managing an interface between a pair of processing cores of a device that are configured to exchange data. The device is configured to enable or disable one or more of the pair of processing cores. One of the methods includes configuring a connect/disconnect interface implemented as logic circuitry between the pair of processing cores to assume a connected state in which the pair of processing cores and exchange data, and configuring the connect/disconnect interface between the pair of processing cores to assume a disconnected state in which one or more of the processing cores is unable to receive data.
A waveguide includes an outcoupler that is implemented in the waveguide as a set of reflective facets that is arranged along a first direction. Each reflective facet is made by applying a reflective coating to a planar face of one or more substrates. Adjacent reflective facets in the set of reflective facets overlap one another along the first direction. For example, a leading portion of one reflective facet in the set of reflective facets overlaps with a tailing portion of the reflective facet adjacent to it.
A method for accessing localized services of a hosting network is implemented in a user equipment (UE) associated with a home network. The method includes receiving an indication of whether the UE is to access the localized services of the hosting network via the hosting network or a serving network distinct from the hosting network (1212); and accessing the localized services in accordance with the indication (i) directly via a radio access network (RAN) of the hosting network (1240) or (ii) via a RAN of the serving network operating as an underlay network, and the hosting network operating as an overlay network (1242).
A method may receive an image representing displayable content for display by an application. A method may execute a layout extraction model using the image as input and generating a list of elements for the image as output, the list of elements including at least a bounding box defining a portion of the image and a role attribute. A method may add the role attribute to a node in an accessibility tree using the list of elements.
A method (500) includes receiving a sequence of acoustic frames (100) as input to a multilingual automated speech recognition (ASR) model (200) configured to recognize speech in a plurality of different supported languages and generating, by an audio encoder (204) of the multilingual ASR, a higher order feature representation (212, 222) for a corresponding acoustic frame. The method also includes generating, by a language identification (LID) predictor (230) of the multilingual ASR, a language prediction representation (232) for a corresponding higher order feature representation. The method also includes generating, by a decoder (240) of the multilingual ASR, a probability distribution (252) over possible speech recognition results based on the corresponding higher order feature representation, a sequence of non-blank symbols (121), and a corresponding language prediction representation. The decoder includes monolingual output layer (400) having a plurality of output nodes (410) each sharing a plurality of language-specific wordpiece models (420).
A method (600) includes obtaining a multi-utterance training sample (410) that includes audio data (412) characterizing utterances spoken by two or more different speakers (10) and obtaining ground-truth speaker change intervals (414) indicating time intervals in the audio data where speaker changes among the two or more different speakers occur. The method also includes processing the audio data to generate a sequence of predicted speaker change tokens (302) using a sequence transduction model (300). For each corresponding predicted speaker change token, the method includes labeling the corresponding predicted speaker change token as correct when the predicted speaker change token overlaps with one of the ground-truth speaker change intervals. The method also includes determining a precision metric (442) of the sequence transduction model based on a number of the predicted speaker change tokens labeled as correct and a total number of the predicted speaker change tokens.
G10L 17/04 - Training, enrolment or model building
G10L 15/06 - Creation of reference templates; Training of speech recognition systems, e.g. adaptation to the characteristics of the speaker's voice
G06N 3/0442 - Recurrent networks, e.g. Hopfield networks characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
A device, method and article of manufacture related to a mechanism for cross profile intent resolution is disclosed. -An example method includes, in response to detecting a first user input associated with a first user profile stored on the computing device, generating a first intent that corresponds to the first user input, applying a sequence of cross profile intent filters to traverse a user profile hierarchy from the first user profile to a second user profile stored on the computing device, wherein the traversal of the user profile hierarchy is based on a successful resolution of each cross profile intent filter of the sequence of cross profile intent filters, identifying an application associated with the second user profile and configured to satisfy the first intent, and providing functionality from the application to satisfy the first intent via the first user profile.
This document describes systems and techniques for a customizable user interface for a device management system. In aspects, a user interface of a device management system includes one or more widgets grouped by at least one category. Each widget of the one or more widgets is associated with at least one network-connected device and is configured to provide at least one of an action functionality, an automation functionality, or image data. Widgets can be organized within spaces to enhance user experience.
G06F 3/0481 - Interaction techniques based on graphical user interfaces [GUI] based on specific properties of the displayed interaction object or a metaphor-based environment, e.g. interaction with desktop elements like windows or icons, or assisted by a cursor's changing behaviour or appearance
G06F 3/0484 - Interaction techniques based on graphical user interfaces [GUI] for the control of specific functions or operations, e.g. selecting or manipulating an object, an image or a displayed text element, setting a parameter value or selecting a range
G06F 3/04847 - Interaction techniques to control parameter settings, e.g. interaction with sliders or dials
H04L 12/28 - Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
H04N 23/661 - Transmitting camera control signals through networks, e.g. control via the Internet
H05B 47/19 - Controlling the light source by remote control via wireless transmission
This document describes systems and techniques for an enhanced video-playback interface. In aspects, a first region displays a first set of images including at least one image, a horizontal timeline, and a horizontal time indicator configured to transition with respect to the horizontal timeline. A second region displays a vertical timeline and a vertical time indicator on the vertical timeline configured to transition with respect to the vertical time indicator. The horizontal time indicator or the vertical timeline can be transitioned with respect to the horizontal timeline or the vertical time indicator, respectively, causing the first region to display a second set of images. In this way, the enhanced video-playback interface can provide an overview of events captured by a camera and enable low-resolution or high-resolution scrubbing through images in sets of image data.
G06F 3/0481 - Interaction techniques based on graphical user interfaces [GUI] based on specific properties of the displayed interaction object or a metaphor-based environment, e.g. interaction with desktop elements like windows or icons, or assisted by a cursor's changing behaviour or appearance
G06F 3/0484 - Interaction techniques based on graphical user interfaces [GUI] for the control of specific functions or operations, e.g. selecting or manipulating an object, an image or a displayed text element, setting a parameter value or selecting a range
G06F 3/04847 - Interaction techniques to control parameter settings, e.g. interaction with sliders or dials
H04L 12/28 - Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
H04N 23/661 - Transmitting camera control signals through networks, e.g. control via the Internet
H05B 47/19 - Controlling the light source by remote control via wireless transmission
31.
CAMERA SYSTEM INCLUDING A MONOCHROME CAMERA AND A COLOR CAMERA HAVING GLOBAL SHUTTER SENSORS
A camera system includes a monochrome camera, having a global shutter, to capture a first image of a scene, and a color camera, disposed separately from the monochrome camera and having a global shutter, to capture a second image of the scene. The second image is aligned to the first image and color information of the second image is provided to the first image to obtain a third image representing the scene.
H04N 23/45 - Cameras or camera modules comprising electronic image sensors; Control thereof for generating image signals from two or more image sensors being of different type or operating in different modes, e.g. with a CMOS sensor for moving images in combination with a charge-coupled device [CCD] for still images
H04N 23/951 - Computational photography systems, e.g. light-field imaging systems by using two or more images to influence resolution, frame rate or aspect ratio
G06T 5/50 - Image enhancement or restoration by the use of more than one image, e.g. averaging, subtraction
32.
INDICATION OF CONFIDENCE IN MACHINE-LEARNED OUTPUTS VIA HAPTIC FEEDBACK
Systems and methods for indicating confidence in a machine-learned output via haptic feedback are provided. For example, a method includes obtaining, by a user computing device, an output of a machine-learned model and an associated confidence metric. The confidence metric is indicative of a degree of confidence in the output of the machine-learned model. The method includes determining a haptic feedback signal indicative of the confidence metric. The method includes receiving data indicative of an input associated with the output of the machine-learned model by a user of the user computing device. The method includes, responsive to receiving the data indicative of the input associated with the output of the machine-learned model, causing performance of the haptic feedback signal for the user via one or more haptic feedback devices.
Example embodiments relate to field of view correction techniques for shutterless camera systems. A mobile device displaying an initial preview of a scene being captured by an image capturing device of the computing device may determine a zoom operation configured to cause the imaging capturing device to focus on a target. The imaging capturing device is configured to change focal length when performing the zoom operation. While the image capturing device performs the zoom operation, the computing device may then map focal lengths used by the imaging capturing device to a virtual focal length such that a field of view of the scene remains consistent across image frames displayed by the display screen between the initial preview of the scene and the zoomed preview of the scene that focuses on the target and display the zoomed preview of the scene that focuses on the target.
H04N 23/63 - Control of cameras or camera modules by using electronic viewfinders
H04N 23/67 - Focus control based on electronic image sensor signals
H04N 23/68 - Control of cameras or camera modules for stable pick-up of the scene, e.g. compensating for camera body vibrations
H04N 23/69 - Control of means for changing angle of the field of view, e.g. optical zoom objectives or electronic zooming
G02B 7/04 - Mountings, adjusting means, or light-tight connections, for optical elements for lenses with mechanism for focusing or varying magnification
G02B 7/28 - Systems for automatic generation of focusing signals
A method (500) includes receiving an input audio signal (122) that corresponds to utterances (120) spoken by multiple speakers. The method also includes processing the input audio to generate a transcription (200) of the utterances and a sequence of speaker turn tokens (224) each indicating a location of a respective speaker turn. The method also includes segmenting the input audio signal into a plurality of speaker segments (225) based on the sequence of speaker tokens. The method also includes extracting a speaker-discriminative embedding from each speaker segment and performing spectral clustering on the speaker-discriminative embeddings to cluster the plurality of speaker segments into k classes. The method also includes assigning a respective speaker label (250) to each speaker segment clustered into the respective class that is different than the respective speaker label assigned to the speaker segments clustered into each other class of the k classes.
An example folding device includes a hinge assembly that is coplanar with the continuous display of the device in order to decrease the thickness of the device. The hinge assembly includes torque members that increase the amount of force needed to rotate the assemblies. In this way, the torque members may provide the device with a more rigid feel. Also in this way, the torque members may enable the device to hold intermediate positions between fully open and fully closed.
Various arrangements are presented that include a pair of true wireless earbuds. The second earbud can be configured to receive an audio packet addressed to only the first earbud and outputs audio based on the audio packet. This audio packet, despite being transmitted by the audio source to only the first earbud, can include audio data for two audio channels.
Implementations described herein identify and correct automatic speech recognition (ASR) misrecognitions. For example, on-device processor(s) of a client device may generate a predicted textual segment that is predicted to correspond to spoken utterance of a user of the client device, and may receive further input that modifies the predicted textual segment to an alternate textual segment. Further, the on-device processor(s) may store these textual segments in on-device storage as a candidate correction pair, and transmit the candidate correction pair to a remote system. Moreover, remote processor(s) of the remote system may determine that the candidate correction pair is an actual correction pair, and may cause client devices to generate updates for a global ASR model for the candidate correction pair. Additionally, the remote processor(s) may distribute the global ASR model to the client devices and/or additional client devices.
Various arrangements for short-range wireless communication between audio output devices, such as true wireless earbuds, are presented herein. A first earbud of a pair of earbuds may determine that a first audio packet addressed to the first earbud from an audio source was not properly received. However, a second earbud of the pair of earbuds may properly receive the first audio packet addressed to the first earbud. The second earbud can then, directly to the first earbud, transmit a cross acknowledgement indicating that the second earbud properly received the audio packet.
A method (500) includes receiving a first query (116) issued by a first user, the first query including a command (111) for a digital assistant (105) to perform a first action, and enabling a round robin mode (350) to control performance of actions. The method also includes, while performing the first action, receiving audio data (402) corresponding to a second query (146) including a command to perform a second action, performing speaker identification on the audio data, determining that the second query was spoken by the first user, preventing performing the second action, and prompting at least another user to issue a query. The method further includes receiving a third query (148) issued by a second user, the third query including a command for the digital assistant to perform a third action, and when the digital assistant completes performing the first action, executing performance of the third action.
An example method includes displaying a zoomed preview of a scene captured by a camera system. The method includes determining a phase-detect autofocus (PDAF) depth estimate and a time-of-flight (ToF) depth estimate for the scene. The method includes, based on a comparison of the PDAF and ToF depth estimates, determining whether a foreground object in the zoomed preview is in-focus for a ToF based AF mode. The method includes, based on a determination that the foreground object in the zoomed preview is in-focus for the ToF based AF mode, bypassing a PDAF mode and activating the ToF based AF mode to focus on the foreground object. The method includes displaying, based on the ToF based AF mode, the focused foreground object as part of the zoomed preview of the scene.
H04N 23/67 - Focus control based on electronic image sensor signals
H04N 23/667 - Camera operation mode switching, e.g. between still and video, sport and normal or high and low resolution modes
H04N 23/69 - Control of means for changing angle of the field of view, e.g. optical zoom objectives or electronic zooming
H04N 23/959 - Computational photography systems, e.g. light-field imaging systems for extended depth of field imaging by adjusting depth of field during image capture, e.g. maximising or setting range based on scene characteristics
G02B 7/28 - Systems for automatic generation of focusing signals
G02B 7/34 - Systems for automatic generation of focusing signals using different areas in a pupil plane
G02B 7/36 - Systems for automatic generation of focusing signals using image sharpness techniques
G02B 7/40 - Systems for automatic generation of focusing signals using time delay of the reflected waves, e.g. of ultrasonic waves
A method (600) for handling contradictory queries on a shared device includes receiving a first query (106) issued by a first user (106a), the first query specifying a first long-standing operation (111) for a digital assistant (105) to perform, and while the digital assistant is performing the first long-standing operation, receiving a second query (146), the second query specifying a second long-standing operation (112) for the digital assistant to perform. The method also includes determining that the second query was issued by another user (102b) different than the first user and determining, using a query resolver (340), that performing the second long-standing operation would conflict with the first long-standing operation. The method further includes identifying one or more compromise operations (354) for the digital assistant to perform, and instructing the digital assistant to perform a selected compromise operation among the identified one or more compromise operations.
Techniques are described herein for table cell splitting in an online document editor. A method includes: responsive to a request to split a cell in a table, determining a target number of rows and a target number of columns, automatically inserting rows adjacent to rows of the cell to reach the target number of rows, automatically inserting columns adjacent to columns of the cell to reach the target number of columns, and automatically merging groups of cells within an initial boundary of the cell, each group spanning a determined number of rows per group and a determined number of columns per group.
A method includes receiving an image frame captured by an image capturing device. The method also includes determining a saliency heatmap representing saliency of pixels in the image frame. The method further includes determining, based on the saliency heatmap, a primary region of interest (ROI) and a secondary ROI for the image frame. The method additionally includes determining a filtered ROI for the image frame, where the filtered ROI updates from a previous filtered ROI to the primary ROI or the secondary ROI based on a saliency difference between the previous filtered ROI and the primary ROI or the secondary ROI exceeding a first threshold. The method also includes applying one or more auto-focus processes based on the filtered ROI, the primary ROI, or the secondary ROI.
This document describes systems and techniques directed to providing real-time feedback to improve self-portrait photographs (selfies) or other images for camera users (e.g., low-vision camera users). In aspects, the systems and techniques are implemented on computing devices having a front-facing camera or a rear-facing camera. The systems and techniques may track the user's face and provide haptic, audio, and/or visual feedback to guide the user to position at least one of the computing device or the user so that the user becomes positioned in a center of frame of the camera. In an aspect, a user interface may display visual indicators that flash over a viewfinder image of a camera displayed on a computing device display. The visual indicators may increase in brightness near a user's face to guide a user to the center of the frame of the camera. In another aspect, the user interface may display a high-contrast outline of the user's face and/or torso on the display to provide feedback to the user of their position in the frame. In another aspect, the user may receive an audio detail description of what is in the viewfinder to confirm desired faces and objects are included. Through such systems and techniques, a user can take a high-quality self-portrait even when they have limited or no ability to see a display screen of the computing device.
H04N 23/61 - Control of cameras or camera modules based on recognised objects
H04N 23/611 - Control of cameras or camera modules based on recognised objects where the recognised objects include parts of the human body
H04N 23/63 - Control of cameras or camera modules by using electronic viewfinders
H04N 23/45 - Cameras or camera modules comprising electronic image sensors; Control thereof for generating image signals from two or more image sensors being of different type or operating in different modes, e.g. with a CMOS sensor for moving images in combination with a charge-coupled device [CCD] for still images
G06V 40/60 - Static or dynamic means for assisting the user to position a body part for biometric acquisition
G06F 3/01 - Input arrangements or combined input and output arrangements for interaction between user and computer
G06F 3/048 - Interaction techniques based on graphical user interfaces [GUI]
This document describes power supplies for compute cores. In one aspect, a power supply system for a compute core includes a primary power converter configured to provide and regulate direct current (DC) power to the compute core over a power rail that electrically couples an output of the primary power converter to the compute core. The power supply system also includes a transient suppressor circuit coupled to the power rail and configured to suppress transient voltage differences between a target supply voltage for the compute core and an actual supply voltage to the compute core.
H02M 3/158 - Conversion of dc power input into dc power output without intermediate conversion into ac by static converters using discharge tubes with control electrode or semiconductor devices with control electrode using devices of a triode or transistor type requiring continuous application of a control signal using semiconductor devices only with automatic control of output voltage or current, e.g. switching regulators including plural semiconductor devices as final control devices for a single load
H02M 1/00 - APPARATUS FOR CONVERSION BETWEEN AC AND AC, BETWEEN AC AND DC, OR BETWEEN DC AND DC, AND FOR USE WITH MAINS OR SIMILAR POWER SUPPLY SYSTEMS; CONVERSION OF DC OR AC INPUT POWER INTO SURGE OUTPUT POWER; CONTROL OR REGULATION THEREOF - Details of apparatus for conversion
An example method includes displaying, by a display screen of an image capturing device, a preview of an image representing a field of view of the image capturing device. The method includes determining a region of interest in the preview. The method includes transitioning the image capturing device from a normal mode of operation to a zoomed mode of operation. The zoomed mode of operation includes: determining, based on sensor data collected by a sensor associated with the image capturing device, a motion trajectory for the region of interest, and based on the determined motion trajectory, generating an adjusted preview representing a zoomed portion of the field of view. The adjusted preview displays the region of interest at or near a center of the zoomed portion. The method includes providing the adjusted preview of the portion of the field of view.
Improved multi-stage methods for training models to enhance input images are provided. The multi-stage methods include training a first model to predict high-quality images based on synthetically degraded versions thereof. The first model is then used to generate, from the high quality images, enhanced, images that can then be used (in combination with synthetically degraded versions thereof) to train additional image enhancement models at two different resolutions. The additional image enhancement models are then applied, in series, to enhance input images. Such a serial image enhancement pipeline can then be used to train a smaller student model that can be implemented on smartphones or other limited-resource systems. This can include using the serial image enhancement pipeline to generate enhanced versions of low-quality images (e.g., as might be generated from a front-facing smartphone camera) that can then be used with the input low-quality images to train the student model.
On-device processor(s) of a client device may store, in on-device storage and in association with a time to live (TTL) in the on-device storage, a correction directed to ASR processing of audio data. The correction may include a portion of a given speech hypothesis that was modified to an alternate speech hypothesis. Further, the on-device processor(s) may cause an on-device ASR model to be personalized based on the correction. Moreover, and based on additional ASR processing of additional audio data, the on-device processor(s) may store, in the on-device storage and in association with an additional TTL in the on-device storage, a pseudo-correction directed to the additional ASR processing. Accordingly, the on-device processor(s) may cause the on-device ASR model to be personalized based on the pseudo-correction to prevent forgetting by the on-device ASR model.
A computing device engages in text-based validation of a user interface (UI) presented on a display of the computing device, including (i) capturing a screenshot of the display when the UI is presented on the display, (ii) transmitting to a server a validation request providing the captured screenshot, and (Hi) receiving from the server, in response to the validation request, a validation response based at least on (a) character recognition of text depicted by the screenshot and (b) a determination of whether the character-recognized text corresponds with an associated action. Further, the computing device uses the received validation response as a basis to control whether to allow the computing device to take the associated action in response to user input into the computing device when the UI is presented on the display.
G06F 21/32 - User authentication using biometric data, e.g. fingerprints, iris scans or voiceprints
G06F 21/52 - Monitoring users, programs or devices to maintain the integrity of platforms, e.g. of processors, firmware or operating systems during program execution, e.g. stack integrity, buffer overflow or preventing unwanted data erasure
G06F 21/54 - Monitoring users, programs or devices to maintain the integrity of platforms, e.g. of processors, firmware or operating systems during program execution, e.g. stack integrity, buffer overflow or preventing unwanted data erasure by adding security routines or objects to programs
G06F 21/53 - Monitoring users, programs or devices to maintain the integrity of platforms, e.g. of processors, firmware or operating systems during program execution, e.g. stack integrity, buffer overflow or preventing unwanted data erasure by executing in a restricted environment, e.g. sandbox or secure virtual machine
A method (800) for securing messages includes obtaining, at a first message server (160), a message (152) for a user (12) of a message service hosted by the first message server, the message including a header (310) including a digital signature (330) signed by an author of the message and a list of one or more recipients (312) of the message. The method includes determining that a Domain Name System (DNS) TXT record (720) associated with the message includes a delegation policy (722) indicating that a second message server declared all intended recipients of the message. In response, the method includes determining that the digital signature by the author is valid and that the user is a declared recipient of the message. The method includes, in response to determining that the digital signature is valid and the user is the declared recipient of the message, indicating the message is authentic.
H04L 9/32 - Arrangements for secret or secure communications; Network security protocols including means for verifying the identity or authority of a user of the system
H04L 51/212 - Monitoring or handling of messages using filtering or selective blocking
H04L 61/4511 - Network directories; Name-to-address mapping using standardised directory access protocols using domain name system [DNS]
H04L 51/214 - Monitoring or handling of messages using selective forwarding
51.
MACHINE LEARNING MODEL BASED TRIGGERING MECHANISM FOR IMAGE ENHANCEMENT
A method includes determining a respective delta quality score associated with each of a plurality of images by predicting, by an image enhancement model, an enhanced image corresponding to a given image, determining a first quality score associated with the given image and a second quality score associated with the enhanced image. The delta quality score is based on a difference of the first and second quality scores. The method includes generating a training dataset comprising the plurality of images associated with respective delta quality scores. The method includes training, based on the generated training dataset, a quality assessment model to predict a quality-improvability score associated with an input image. The quality-improvability score is indicative of a potential to increase a perceptual quality of the input image based on removal of one or more image degradation factors. The method includes outputting, by the computing device, the trained quality assessment model.
A method includes receiving an image captured by an image capturing device. The method also includes determining a saliency bounding box based on a saliency metric determined for pixels of the image. The method further includes determining one or more face bounding boxes surrounding one or more faces identified within the image. The method additionally includes determining a zoom bounding box based on the saliency bounding box and the one or more face bounding boxes. The method also includes determining a zoom ratio based on the determined zoom bounding box. The method further includes providing a zoomed image for display based on the determined zoom ratio.
This document describes systems and techniques directed at customizable automations for network-connected devices. In aspects, a device management system presents a starter input having a trigger menu and a detecting device menu. The device management system receives user input indicative of a selected trigger and a detecting device from one or more of the menus. The device management system also presents an action input having an action menu and an action device menu. The device management system receives user input indicative of a selected action and a selected action device. Based on the selections, the device management system associates the selected trigger with the selected action such that, responsive to the selected trigger being detected by the selected detecting device, the selected action is performed by the selected action device.
H04L 12/28 - Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
G06F 3/0481 - Interaction techniques based on graphical user interfaces [GUI] based on specific properties of the displayed interaction object or a metaphor-based environment, e.g. interaction with desktop elements like windows or icons, or assisted by a cursor's changing behaviour or appearance
G06F 3/0484 - Interaction techniques based on graphical user interfaces [GUI] for the control of specific functions or operations, e.g. selecting or manipulating an object, an image or a displayed text element, setting a parameter value or selecting a range
G06F 3/04847 - Interaction techniques to control parameter settings, e.g. interaction with sliders or dials
A method in a user equipment (UE) equipped with a first transmitter and a second transmitter, the method comprising: transmitting a first uplink transmission using the first transmitter switched to a first frequency band and using a second transmitter switched to a second frequency band; and receiving, from a radio access network (RAN), an uplink switching configuration for a second uplink transmission using the first transmitter, the uplink switching configuration including (i) a first parameter to indicate whether to switch the second transmitter away from the second frequency band, and (ii) a second parameter indicating to which frequency band the UE is to switch the second transmitter; transmitting the second uplink transmission in accordance with the uplink switching configuration.
This disclosure provides systems, devices, apparatus, and methods, including computer programs encoded on storage media, for UE reports for STxMP. A UE (102) receives (604), from a network entity (104), control signaling for transmission (616) of a report associated with STxMP. The control signaling indicates at least one of a reporting quantity of uplink beams for the report, one or more downlink reference signals to be measured for the report, prohibit timer information for the transmission (616) of the report, or an uplink resource for the transmission (616) of the report. The UE transmits (616), to the network entity (104) based on a triggering condition, the report in conformance with the control signaling. The report corresponds to at least one of a beam report for STxMP or a panel status update report for STxMP.
H04B 7/0404 - Diversity systems; Multi-antenna systems, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas the mobile station comprising multiple antennas, e.g. to provide uplink diversity
H04B 7/06 - Diversity systems; Multi-antenna systems, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station
H04B 7/08 - Diversity systems; Multi-antenna systems, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the receiving station
H04L 5/00 - Arrangements affording multiple use of the transmission path
56.
PHOTOREALISTIC TEXT INPAINTING FOR AUGMENTED REALITY USING GENERATIVE MODELS
Provided are systems and methods that use generative models (e.g., generative adversarial networks) to enable photorealistic text inpainting in augmented reality. One example application of the proposed systems is to perform augmented reality translation. For example, a user can operate an image capture device (e.g., camera, smartphone, etc.) to capture imagery of a real-world scene that includes real-world text (e.g., signage, restaurant menus, etc.). The real-world text can be translated into a different language. Further, the captured imagery can be processed with a machine-learned generative model to produce an augmented image. The augmented image can depict the real-world scene with the real-world text removed. Specifically, because a machine-learned generative model is used, the augmented image can appear significantly more realistic, for example versus an image in which the real-world text has simply been blocked using a box with a single color.
The technology is generally directed to providing a next suggested action based on a first user's request for location information of a second user and the location information of the second user. Location information may be shared between the first and second user after each user authorizes location sharing with specific users. The second user's location information may be provided to the first user in response to a request for the first user. Based on the request from the first user and the location information of the second user, a next suggested action may be automatically determined and provided to the first or second user. The suggested next action may be for the first user to send a message to the second user, the second user to send a message to the first user or another user, updating a navigation route, providing an update to a scheduled event, etc.
A computing device may implement a method for providing route information regarding a completed or ongoing trip by a user without the user having previously initiated a navigation session. The method may include receiving a query regarding a previous or ongoing trip by a user prior to the user initiating a navigation session; determining an origin for the previous or ongoing trip; obtaining route information for the previous or ongoing trip; generating one or more route attributes associated with the query based at least on the origin for the previous or ongoing trip and the route information for the previous or ongoing trip; and providing a response to the query based at least on the one or more route attributes.
A system of multiple radar-enabled computing devices, along with related techniques, are described in this document. These techniques are employed with this system to coordinate information and operations across multiple radar-enabled computing devices to create a seamless experience. In particular, each computing device of the computing system may have access to stored radar-signal characteristics that enable detection and distinction of users and detection and recognition of gestures. Computing devices may coordinate in-progress operations to provide continuity across multiple devices. When positioned in different locations, each device may also learn over time users, gestures, and versions of gestures associated with that location to anticipate them in the future.
G06F 3/01 - Input arrangements or combined input and output arrangements for interaction between user and computer
G01S 7/41 - RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES - Details of systems according to groups , , of systems according to group using analysis of echo signal for target characterisation; Target signature; Target cross-section
G01S 13/06 - Systems determining position data of a target
G01S 13/42 - Simultaneous measurement of distance and other coordinates
G01S 13/88 - Radar or analogous systems, specially adapted for specific applications
Techniques and devices for ambiguous gesture determination using contextual information are described in this document for radar-enabled computing devices. Contextual information may include a status of operations that are performed by the radar-enabled computing device or an associated device at a current time, past time, or future time. Contextual information may also or instead include foreground and background operations, a history of operations saved to a memory, scheduled or anticipated operations, a location of a user or device, room-related context, user habits, and so forth. Two or more computing devices may coordinate this contextual information across a communication network to form a computing system.
G06F 3/01 - Input arrangements or combined input and output arrangements for interaction between user and computer
G01S 7/41 - RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES - Details of systems according to groups , , of systems according to group using analysis of echo signal for target characterisation; Target signature; Target cross-section
G01S 13/06 - Systems determining position data of a target
G01S 13/42 - Simultaneous measurement of distance and other coordinates
G01S 13/88 - Radar or analogous systems, specially adapted for specific applications
G06V 40/20 - Movements or behaviour, e.g. gesture recognition
Techniques and devices for radar-based gesture determination at long ranges are described in this document. The techniques described herein enable a computing device to detect and recognize gestures at long-range extents of up to eight meters. The computing device of this disclosure does not require the user to perform a gestural command at a specific location, in a specific orientation, contingent upon a wake-up trigger, or at a specific time, enabling the user to freely provide commands whenever and wherever is most convenient. This continual recognition of gestures may be enabled by a machine-learned model, generation of augmented data, and inclusion of negative data.
G06F 3/01 - Input arrangements or combined input and output arrangements for interaction between user and computer
G01S 7/41 - RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES - Details of systems according to groups , , of systems according to group using analysis of echo signal for target characterisation; Target signature; Target cross-section
G01S 13/88 - Radar or analogous systems, specially adapted for specific applications
G01S 13/50 - Systems of measurement based on relative movement of target
This document describes techniques, apparatuses, and systems for sensor capability determination for radar-based computing devices. Through these techniques, gesture-determination devices may be configured with one or more primary sensors to improve a quality of gesture determination. Specifically, capabilities of first and second sensors to sense, and therefore sensed data to be used to detect or recognize a gesture may be determined based on contextual information associated with a region in which the gesture is performed. These capabilities may be compared to determine that the first sensor is more capable. As a result, a device utilizing the first and second sensors to enable gesture determination may be configured such that the first sensor is a primary sensor to be used preferentially over the second sensor to sense the gesture at a current or future time. In doing so, gesture recognition accuracy may be increased in various environments.
G06F 3/01 - Input arrangements or combined input and output arrangements for interaction between user and computer
G01S 7/41 - RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES - Details of systems according to groups , , of systems according to group using analysis of echo signal for target characterisation; Target signature; Target cross-section
G01S 13/06 - Systems determining position data of a target
G01S 13/42 - Simultaneous measurement of distance and other coordinates
G01S 13/88 - Radar or analogous systems, specially adapted for specific applications
G06V 40/20 - Movements or behaviour, e.g. gesture recognition
This document describes techniques, apparatuses, and systems for the determination of a less-destructive command. A computing device may detect an ambiguous gesture performed by a user and compare a radar-signal characteristic of the ambiguous gesture to one or more stored radar-signal characteristics to correlate the ambiguous gesture to a first gesture and a second gesture. The first gesture and the second gesture may cause the computing device to perform a first command and a second command, respectively. The computing device may determine a less-destructive command of the first and second command and perform an operation associated with the less-destructive command. In doing so, a device performing radar-based gesture detection may reduce the consequences of inaccurate gesture recognition, thereby improving user satisfaction.
G06F 3/01 - Input arrangements or combined input and output arrangements for interaction between user and computer
G01S 7/41 - RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES - Details of systems according to groups , , of systems according to group using analysis of echo signal for target characterisation; Target signature; Target cross-section
G01S 13/06 - Systems determining position data of a target
This document describes techniques, apparatuses, and systems for determining user engagement. For example, a computing device may determine a current proximity, a projected proximity, or a body orientation of a user relative to an interaction device associated with the computing device. Using one or more of these determinations, the techniques estimate an engagement or projected engagement of the user with the interaction device. With this estimate, the techniques alter a setting of the interaction device to better interact with the user.
G06F 3/01 - Input arrangements or combined input and output arrangements for interaction between user and computer
G01S 7/41 - RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES - Details of systems according to groups , , of systems according to group using analysis of echo signal for target characterisation; Target signature; Target cross-section
G01S 13/06 - Systems determining position data of a target
G01S 13/42 - Simultaneous measurement of distance and other coordinates
G01S 13/88 - Radar or analogous systems, specially adapted for specific applications
G06V 40/20 - Movements or behaviour, e.g. gesture recognition
65.
IN-LINE LEARNING OF NEW GESTURES FOR RADAR-ENABLED COMPUTING DEVICES
Techniques, apparatuses, and systems for in-line learning of new gestures for radar-enabled computing devices are described in this document. A computing system may store radar-signal characteristics of a new gesture to enable the computing system to recognize a new gesture and perform a command associated with the new gesture. Specifically, a radar system may detect a gesture performed by a user and fail to correlate that gesture to one or more known gestures. The computing system may receive a new command proximate to detecting the gesture and determine that the detected gesture is a new gesture associated with the new command. As such, the computing system may store a radar-signal characteristic of the new gesture effective to recognize a performance of the gesture in the future and respond by performing the command. In doing so, the computing system may periodically learn new gestures without requiring dedicated training from the user.
G06F 3/01 - Input arrangements or combined input and output arrangements for interaction between user and computer
G01S 7/41 - RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES - Details of systems according to groups , , of systems according to group using analysis of echo signal for target characterisation; Target signature; Target cross-section
G01S 13/06 - Systems determining position data of a target
G01S 13/42 - Simultaneous measurement of distance and other coordinates
G01S 13/88 - Radar or analogous systems, specially adapted for specific applications
G06V 40/20 - Movements or behaviour, e.g. gesture recognition
66.
PRESENTING RELATED CONTENT WHILE BROWSING AND SEARCHING CONTENT
Systems and methods for presenting an interface for additional content suggestion can include obtaining data descriptive of the displayed content and determining additional content associated with the displayed content. An interface can then be provided that displays data associated with the displayed content and the additional content. The interface can include a first viewing window for displaying a portion of the displayed content and a second viewing window for displaying a snippet associated with the additional content.
Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating an output image using a text-to-image model and conditioned on both the input text and image and text pairs selected from a multi-modal knowledge base. In one aspect, a method includes, at each of multiple time steps: generating a first feature map for the time step; selecting one or more neighbor image and text pairs based on their similarities to the input text; for each of the one or more neighbor images and text pairs, generating a second feature map for the neighbor image and text pair; applying an attention mechanism over the one or more second feature maps to generate an attended feature map; and generating an updated intermediate representation of the output image for the time step.
G06F 16/783 - Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
A method (300) for a crawl algorithm includes obtaining a plurality of web pages (152) for a web crawler (160) to crawl. The method also includes determining an available bandwidth (155) for the web crawler. The method includes, for each respective web page of the plurality of web pages, determining a respective crawl value (153) for the respective web page based on the available bandwidth and determining that the respective crawl value of the respective web page satisfies a threshold value (162). The method includes, in response to determining that the respective crawl value of the respective web page satisfies the threshold value, updating the respective web page in a cache memory (150).
This document describes systems and techniques for removing distortion from real-time video using a masked frame. In aspects, an image-capture device having a video-processing manager is configured to capture a video segment comprising a sequence of frames. The sequence of frames includes at least a current frame having a foreground and a background. The video-processing manager receives a subject mask, motion vectors, and a predicted mask for the current frame. The video-processing manager generates a final mask for the current frame based on the subject mask, motion vectors, and predicted mask. The video-processing manager applies the final mask to the current frame to segment the foreground from the background and provide a masked frame. The video-processing manager edits the masked frame to remove distortion to generate an output frame and outputs the output frame. By repeating the method described for each frame in the sequence of frames, the video-processing manager provides an improved video segment.
Aspects of the disclosure provide a deep sequence model, referred to as Koopman Neural Forecaster (KNF), for time series forecasting. KNF leverages deep neural networks (DNNs) to learn the linear Koopman space and the coefficients of chosen measurement functions. KNF imposes appropriate inductive biases for improved robustness against distributional shifts, employing both a global operator to learn shared characteristics, and a local operator to capture changing dynamics, as well as a specially-designed feedback loop to continuously update the learnt operators over time for rapidly varying behaviors. KNF achieves superior performance on multiple time series datasets that are shown to suffer from distribution shifts.
Provided is a noisy channel generative model of two sequences, for example text and speech, which enables uncovering the associations between the two modalities when limited paired data is available. To address the intractability of the exact model under a realistic data set-up, example aspects of the present disclosure include a variational inference approximation. To train this variational model with categorical data, a KL encoder loss approach is proposed which has connections to the wake-sleep algorithm.
G06N 3/088 - Non-supervised learning, e.g. competitive learning
G10L 13/08 - Text analysis or generation of parameters for speech synthesis out of text, e.g. grapheme to phoneme translation, prosody generation or stress or intonation determination
G10L 15/16 - Speech classification or search using artificial neural networks
72.
MANAGING PDCP OPERATION IN A SERVING CELL CHANGE SCENARIO
A node of a radio access network (RAN) transmits, to a user equipment (UE) communicating with the RAN in a first cell and using a radio bearer, a message including a configuration for performing a serving cell change to a second cell subsequent to an activation command, including refraining from including a Packet Data Convergence Protocol (PDCP) reestablishment indication in the message; and transmits, to the UE and subsequent to the transmitting of the message including the configuration, an activation command for performing the serving cell change to the second cell in accordance with the configuration and without reestablishing a PDCP entity of the UE for the radio bearer.
A node in a RAN transmits (1007) to a user equipment UE in a first cell, a message including a configuration for performing a serving cell change to a second cell subsequent to an activation command; determines (1008), subsequent to the transmitting and while the UE awaits the activation command, a communication failure between the UE and the RAN; and in response to the determining, releases (1011, 1013) the configuration.
This disclosure provides methods for selecting a precoder for uplink sounding reference signal (SRS), useful in multi-transmission and reception point (multi-TRP) cases. A user equipment (UE) device receives (450), from at least a first network entity, a first channel state information reference signal (CSI-RS), and also receives (455), from a second network entity, a second CSI-RS. The UE device then receives (430), from the first network entity, for example, control signaling indicating at least one SRS resource set associated with the first CSI-RS or the second CSI-RS. The UE device transmits (470) precoded SRS resources based on the received SRS resource set and a precoder computed based on the first CSI-RS and the second CSI-RS. In some cases, the UE device determines (460) the precoder based on the first CSI-RS and the second CSI-RS such that the precoded SRS resources suppress interference with signals received at a third network entity.
H04B 7/024 - Co-operative use of antennas at several sites, e.g. in co-ordinated multipoint or co-operative multiple-input multiple-output [MIMO] systems
H04B 7/0456 - Selection of precoding matrices or codebooks, e.g. using matrices for antenna weighting
H04B 7/06 - Diversity systems; Multi-antenna systems, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station
H04L 5/00 - Arrangements affording multiple use of the transmission path
This disclosure provides systems, devices, apparatus, and methods, including computer programs encoded on storage media, for interference-aware uplink power Control. A UE (302) receives (306) a first control signal indicating a plurality of power control parameter sets. The UE (302) receives (308) a second control signal to trigger an uplink signal based on at least one of the plurality of power control parameter sets. The UE (302) transmits (314) the uplink signal with a transmission power determined based on the at least one of the plurality of power control parameter sets.
H04W 52/36 - Transmission power control [TPC] using constraints in the total amount of available transmission power with a discrete range or set of values, e.g. step size, ramping or offsets
A UE (102) receives (310a-310b), from a network entity (104), at least one downlink signal for monitoring (316a-316b) a performance of an ML model used for CSI compression. The monitoring (316a-316b) the performance is based on a measurement value of the at least one downlink signal. The UE transmits (318, 414a-414b) to the network entity (104) based on the measurement value of the at least one downlink signal, information associated with the performance of the ML model used for the CSI compression. The UE communicates (322) with the network entity (104) when the information associated with the performance of the ML model indicates a performance failure, where the communication (322) applies at least one of: an update to the ML model, a switch of the ML model to a different ML model, or non-ML CSI reporting.
This disclosure provides systems, devices, apparatus, and methods, including computer programs encoded on storage media, for UE-triggered TDCP reports. A UE (102) receives (208), from a network entity (104), control signaling that indicates a triggering condition for the TDCP report. The triggering condition is associated with a measurement value of at least one downlink reference signal. The UE (102) receives (210a-210b), from the network entity (104), the at least one downlink reference signal, where the measurement value of the at least one downlink reference signal can correspond to detection of the triggering condition for the TDCP report. The UE (102) transmits (218) to the network entity (104), and the network entity (104) receives (218) from the UE (102), the TDCP report based on the detection of the triggering condition for the TDCP report.
This disclosure provides systems, devices, apparatus, and methods, including computer programs encoded on storage media, for DMRS configuration. A UE receives (904) control signaling that causes the UE to enable at least one of: a number of eType1 or eType2 DMRS antenna ports, a minimal FD-OCC de-spreading length, or at least one orphan RE handling scheme. The UE receives (306) DCI that schedules a physical shared channel for at least one of: at least one indicated eType1/eType2 DMRS antenna port, an indicated FD-OCC de-spreading length, or an indicated orphan RE handling scheme. The UE communicates (911) with the network entity on the physical shared channel based on at least one of: the at least one indicated eType1/eType2 DMRS antenna port, the indicated FD-OCC de-spreading length, or the indicated orphan RE handling scheme.
Methods, systems, and apparatus, including medium-encoded computer program products, for adaptive content distribution using private encoded audio identifiers are described. The techniques can include receiving event data that indicates that a digital component with an audio signature was transmitted to a display device. The event data can also include a time at which the digital component was transmitted. A content request can be received from a different client device and can include data representative of a captured audio signature and the time at which the audio signature was captured. In response to determining that the content request is requesting content related to the digital component based at least on (i) a determination that the audio signature matches the audio signature of the digital component and (ii) a determination that the time are within a threshold duration, the content related to the digital component can be sent to the client device.
H04N 21/2389 - Multiplex stream processing, e.g. multiplex stream encrypting
H04N 21/439 - Processing of audio elementary streams
G10L 19/018 - Audio watermarking, i.e. embedding inaudible data in the audio signal
H04N 21/258 - Client or end-user data management, e.g. managing client capabilities, user preferences or demographics or processing of multiple end-users preferences to derive collaborative data
H04N 21/45 - Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies
H04N 21/8547 - Content authoring involving timestamps for synchronizing content
H04N 21/41 - Structure of client; Structure of client peripherals
H04N 21/4722 - End-user interface for requesting content, additional data or services; End-user interface for interacting with content, e.g. for content reservation or setting reminders, for requesting event notification or for manipulating displayed content for requesting additional data associated with the content
H04N 21/442 - Monitoring of processes or resources, e.g. detecting the failure of a recording device, monitoring the downstream bandwidth, the number of times a movie has been viewed or the storage space available from the internal hard disk
H04N 21/422 - Input-only peripherals, e.g. global positioning system [GPS]
An index of a motion vector candidate of a list of motion vector candidates is decoded from a compressed bitstream. A subset of motion vector candidates to generate is determined based on the index. The subset of motion vector candidates is then generated. The subset of motion vector candidates is a proper subset of the list of motion vector candidates. That is, fewer than all of the motion vector candidates of the list of motion vector candidates are generated. The motion vector candidate is selected from the subset of motion vector candidates based on the index. A current block is decoded using the motion vector candidate.
Techniques and devices for user distinction for radar-based gesture detectors are described in this document. These techniques enable a computing device to distinguish users using a radar system that may collect and analyze radar characteristics of a user to distinguish that user from other users. The radar characteristics may include radar-reflection features of the user such as topological, temporal, gestural, and/or contextual information. A user may be distinguished without determining personally identifiable information, and the computing device may record radar characteristics to distinguish each user at a later time and provide tailored experiences. When an unregistered person is detected, the radar system may assign the unregistered person an unregistered user identification that contains detected radar characteristics to distinguish this person from other users at a future time.
G06F 3/01 - Input arrangements or combined input and output arrangements for interaction between user and computer
G01S 7/41 - RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES - Details of systems according to groups , , of systems according to group using analysis of echo signal for target characterisation; Target signature; Target cross-section
G01S 13/06 - Systems determining position data of a target
G01S 13/42 - Simultaneous measurement of distance and other coordinates
G01S 13/88 - Radar or analogous systems, specially adapted for specific applications
G06V 40/20 - Movements or behaviour, e.g. gesture recognition
82.
CONTINUAL IN-LINE LEARNING FOR RADAR-BASED GESTURE RECOGNITION
Techniques and devices for continual in-line learning for radar-based gesture recognition are described in this document. Through continual in-line learning, a computing device may improve recognition of even the hardest-to-recognize gestures by gradually storing characteristics of ambiguous gestures performed by a user. Specifically, a radar system may detect a first ambiguous gesture that the computing device fails to recognize as a known gesture and a second gesture that the computing device successfully recognizes as the known gesture. The computing device may identify a similarity between the first and the second gesture, and in doing so, store a characteristic of the first gesture to recognize the known gesture more-accurately in a future occurrence.
G06F 3/01 - Input arrangements or combined input and output arrangements for interaction between user and computer
G01S 7/41 - RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES - Details of systems according to groups , , of systems according to group using analysis of echo signal for target characterisation; Target signature; Target cross-section
G01S 13/06 - Systems determining position data of a target
G01S 13/42 - Simultaneous measurement of distance and other coordinates
G01S 13/88 - Radar or analogous systems, specially adapted for specific applications
G06V 40/20 - Movements or behaviour, e.g. gesture recognition
This document describes techniques and devices for in-line learning based on user inputs. Through in-line learning, a computing device may store characteristics of ambiguous gestures based on subsequent commands from a user. For example, the ambiguous gesture may be associated to one or more known gestures, but the ambiguous gesture cannot be recognized as one of the known gestures with sufficient confidence for gesture recognition. When the computing device fails to recognize the ambiguous gesture, the user may perform or request the performance of a command. This command may be determined to be a first command associated with a first gesture of the known gestures with which the ambiguous gesture was associated. As such, the computing device may store a characteristic of the ambiguous gesture with the first gesture to improve recognition of the first gesture in the future.
G06F 3/01 - Input arrangements or combined input and output arrangements for interaction between user and computer
G01S 7/41 - RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES - Details of systems according to groups , , of systems according to group using analysis of echo signal for target characterisation; Target signature; Target cross-section
G01S 13/06 - Systems determining position data of a target
G01S 13/42 - Simultaneous measurement of distance and other coordinates
G01S 13/88 - Radar or analogous systems, specially adapted for specific applications
G06V 40/20 - Movements or behaviour, e.g. gesture recognition
84.
SMOOTH CONTINUOUS ZOOMING IN A MULTI-CAMERA SYSTEM BY IMAGE-BASED VISUAL FEATURES AND OPTIMIZED GEOMETRIC CALIBRATIONS
An example method includes displaying an initial preview of a scene being captured by a first camera operating within a first range of focal lengths. The method includes detecting a zoom operation predicted to cause the first camera to reach a limit of the first range. The method includes activating a second camera, operating within a second range of focal lengths, to capture a zoomed preview of the scene. The method includes updating a geometry-based warping transformation based on a comparison of respective image features from the initial preview and the zoomed preview. The method includes aligning the zoomed preview with the initial preview by applying the updated warping transformation. The method includes displaying the aligned zoomed preview of the image captured by the second camera while operating within the second range.
H04N 23/63 - Control of cameras or camera modules by using electronic viewfinders
H04N 23/45 - Cameras or camera modules comprising electronic image sensors; Control thereof for generating image signals from two or more image sensors being of different type or operating in different modes, e.g. with a CMOS sensor for moving images in combination with a charge-coupled device [CCD] for still images
H04N 23/667 - Camera operation mode switching, e.g. between still and video, sport and normal or high and low resolution modes
H04N 23/69 - Control of means for changing angle of the field of view, e.g. optical zoom objectives or electronic zooming
85.
METHODS AND DEVICES FOR HANDLING INTER-FREQUENCY MEASUREMENTS ON NEIGHBORING NTN CELLS
A user equipment (102) is configured to conduct (1218A, 1318) an intra-frequency measurement on a candidate frequency of a non-terrestrial network (NTN) cell even when the UE does not receive information about measurement timing within a candidate frequency configuration or pertinent satellite ephemeris information via a neighbor NTN cell configuration. The UE (102) operates based on the assumption that a base station (104) preparing lists of frequency configurations and neighbor NTN cell configurations has reduced redundant information (e.g., when the same satellite provides plural non-terrestrial cells). The base station (104) is configured to prepare (1501, 1503, 1505, 1605) and broadcast (1508, 1510, 1608, 1610) such reduced lists of frequency configurations and neighboring NTN cell configurations providing information for the intra-frequency measurements.
Aspects of the disclosure are directed to a canonical approach for feature selection referred to as sparse learnable masks (SLM). SLM integrates learnable sparse masks into end-to-end training. For the fundamental non-differentiability challenge of selecting a desired number of features, SLM includes dual mechanisms for automatic mask scaling by achieving a desired feature sparsity and gradually tempering this sparsity for effective learning. SLM further employs an objective that increases mutual information (MI) between selected features and labels in an efficient and scalable manner. Empirically, SLM can achieve or improve upon state-of-the-art results on several benchmark datasets, often by a significant margin, while reducing computational complexity and cost.
Implementations relate to selecting a particular device, from an ecosystem of devices, to provide responses to a device-agnostic request of the user while a scenario is occurring. The user specifies a scenario and contextual features are identified from one or more devices of the ecosystem to generate scenario features indicative of the scenario occurring. The scenario features are stored with a correlation to a device that is specified by the user to handle responses while the scenario is occurring. When a subsequent device-agnostic request is received, current contextual features are identified and compared to the scenario features. Based on the comparison, the specified assistant device is selected to respond to the device-agnostic request.
Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for generating a video. In one aspect, a method comprises receiving a first text prompt, using a video generation neural network to generate an initial segment of the video conditioned on the first text prompt, and updating the video for each of one or more update iterations by obtaining an additional text prompt for each update iteration and by using the video generation neural network to generate an additional segment of the video conditioned on the text prompt for the update iteration.
A method (600) for generation of chat applications includes receiving a deployment request (24) requesting deployment of a no-code application (191) generated by a user (12) within a no-code environment (172, 174) to a chat application environment (202, 204). The no-code application includes a trigger condition (510), an action response (520), and a no-code environment graphical user interface (GUI) view (500) based on the action response. The method includes, after receiving the deployment request, receiving an interaction indication (30) indicating that the trigger condition is satisfied. In response to receiving the interaction indication, the method includes executing the action response, translating the no-code environment GUI view into a chat application GUI view (410), and transmitting the chat application GUI view to a user device (12). The chat application GUI view is configured to cause the user device to display the chat application GUI view within the chat application environment.
G06F 8/38 - Creation or generation of source code for implementing user interfaces
H04L 51/046 - Interoperability with other network applications or services
H04L 51/02 - User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail using automatic reactions or user delegation, e.g. automatic replies or chatbot-generated messages
A first node of a RAN communicates with a UE in a first cell according to a first configuration; transmits, to the UE, a message including a second configuration for accessing a second cell subsequent to an activation command; subsequent to the transmitting and while the UE awaits the activation command, transmits a handover message to a second node of the RAN or a CN; and releases the second configuration.
Existing language models (LMs) can excel at some tasks such as question answering, reasoning, and dialog. However, they can sometimes generate unsupported or inaccurate content. Therefore, in the present disclosure, systems and methods are provided for improving the reliability of LMs' generated output. First, systems and methods are provided for editing LMs' generated content based on a machine-learned comparison between the generated content and related evidence snippets, which can be retrieved and extracted using a machine-learned query generation model and a machine-learned relevance model. Second, systems and methods are provided for attributing parts of LM-generated content (e.g. factual claims) to related evidence snippets. Thus, the present disclosure can improve the reliability of LM output, both by increasing the factual accuracy of edited content and by allowing a user or computing system to know whether parts of the generated content are supported or contradicted by external evidence.
A CU of a distributed base station, which includes the CU and a distributed unit DU, transmits (1006), to a user equipment UE via the DU and in a first cell, a configuration for performing a serving cell change to a second cell subsequent to an activation command; and in response to receiving (1008, 1009) a DU-to-CU message subsequent to the transmitting of the configuration but prior to receiving an indication that the UE has connected to the second cell, suspends (1010) DE transmissions of data packets to the UE.
A plurality of model portions are determined from a machine-learned model based on at least one criterion. A plurality of local optimization functions are respectively determined for the plurality of model portions. Forward-mode differentiation is performed for each model portion of the plurality of model portions. Performing forward-mode differentiation includes applying a perturbation to outputs of one or more model units of the model portion. Performing forward-mode differentiation includes, based at least in part on the perturbation, determining a gradient of the local optimization function for the model portion. Performing forward-mode differentiation includes modifying one or more parameters of the model portion based on the gradient.
G06N 3/082 - Learning methods modifying the architecture, e.g. adding, deleting or silencing nodes or connections
G06N 3/084 - Backpropagation, e.g. using gradient descent
G06N 3/0442 - Recurrent networks, e.g. Hopfield networks characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
A method including receiving an audio signal including a plurality of audio channels, selecting a first portion of the plurality of audio channels, selecting a second portion of the plurality of audio channels, generating first mixed audio channels by mixing the first portion of the plurality of audio channels with a first time-delayed audio channel, generating second mixed audio channels by mixing the second portion of the plurality of audio channels with a second time-delayed audio channel, and generating an augmented ambisonics model based on the plurality of audio channels, the first mixed audio channels, and the second mixed audio channels.
A equipment (UE) communicates with a RAN in a first cell in accordance with a first configuration. The UE receives, from the RAN, a message including a configuration for performing a serving cell change to a second cell subsequent to an activation command; detects a communication failure with the RAN, prior to receiving the activation command; and in response to the detecting, performing at least one of the following: (i) performing an RRC reestablishment procedure, or releasing the second configuration.
A method in a user equipment (UE) equipped with a plurality of transmitters includes receiving (314), from a radio access network (RAN), an uplink switching configuration that indicates, for a plurality of frequency bands including a first frequency band and a second frequency band, respective priorities. The method also includes determining (334), for an uplink transmission to the RAN and based on the respective priorities, whether to allocate a time resource in a first slot associated with the first frequency band or a second slot associated with the second frequency band, the time resource being for the UE to switch at least one of the plurality of transmitters from the first frequency band to the second frequency band.
A transmission method implemented in a UE comprises generating an indication of a remaining delay budget for uplink (UL) data, transmitting, to a radio access network (RAN), the indication of the remaining delay budget, and transmitting the UL data to the RAN, via one or more UL resources allocated by the RAN.
The present disclosure provides for determining personalized banner placement in relation to content based on probabilistic spatial user engagement. The probabilistic spatial user engagement can be determined based on user input signals, types of content, or a combination of user input signals and types of content. Such determination may be used to identify regions of a page displaying the content where banners may be rendered for maximum user engagement and minimal disruption of the content.
A metering stack for collecting a target sample includes a channel layer spacing a top layer from a bottom layer, where the top, bottom, and channel layers together define a channel. The channel has an inlet end, a main channel portion, a separation portion, and one or more dispensing portions. A vent is defined within the metering stack proximate the separation portion, where the vent allows air to enter the metering stack into the separation portion. The vent has a first wall extending between a first end and a second end, and a curved wall extending between the first end and the second end, with at least a portion of the first wall being closer than the curved wall to the main channel portion and with the first wall being at an angle relative to a main axis of the main channel portion.
Methods, systems, and apparatus, for systems on-a-chip. One system includes a functional component having one or more embedded random-access memories (RAMs), the functional component including a scan memory state machine configured to generate signals for dumping the contents of the one or more embedded RAMs during a scan dump process.