Abstract: Provided are an AR translation processing method and an electronic device, which relate to the technical field of communications. By the method, in a scenario in which an electronic device is used for AR translation, a pose change of the electronic device can be detected in real time, and feature matching can be performed on a plurality of consecutive frames of images acquired by a camera, so that whether to-be-translated text needs to be fully translated or partially translated, or needs not to be translated can be determined based on the pose change of the electronic device and a feature matching result, and therefore a corresponding translation trigger strategy is selected. In this way, repeated translation can be effectively avoided, thereby saving computing resources in the AR translation process and improving the translation efficiency to a particular extent.

Abstract: A method for tracking zero-angle position shift of a moveable mirror used in a LiDAR system is provided. The method comprises obtaining a first dataset based on a first intensity map. The first intensity map is associated with internal reflection pulses of a frame scanned by the LiDAR system. The frame comprises a plurality of scan positions. The internal reflection pulses are formed by scattering or reflecting one or more transmission light pulses at positions internal to a housing of the LiDAR system. The first dataset is a calibration dataset comprising representative intensity values and corresponding positions in the frame. The method further comprises obtaining a second intensity map of another frame at a subsequent time and obtaining a second dataset based on the second intensity map. The method further comprises determining the zero-angle position shift of the moveable mirror based on the first dataset and the second dataset.

Abstract: One or more computing devices, systems, and/or methods are provided. A machine learning model may be trained using a plurality of sets of information. One or more pruning operations may be performed in association with the training to generate a machine learning model with sparse vector representations associated with features of the plurality of sets of information. A request for content associated with a client device may be received. A set of features associated with the request for content may be determined. A plurality of positive signal probabilities associated with a plurality of content items may be determined using the machine learning model based upon one or more sparse vector representations, of the machine learning model, associated with the set of features. A content item may be selected from the plurality of content items for presentation via the client device based upon the plurality of positive signal probabilities.

Abstract: Provided are a direct time of flight (dTOF) sensing module, a terminal device, and a ranging method, to resolve a problem that in a conventional technology, a dTOF sensing module cannot adapt to different detection scenarios to some extent. Every K light sensitive units (401) of W light sensitive units (401) included in the dTOF sensing module share a first storage space. A processing control unit (403) is configured to control gating of N light sensitive units (401). The gated N light sensitive units (401) occupy the first storage space, and Q time slice bins are allocated to each gated light sensitive unit (401), so that the dTOF sensing module operates in a first mode or a second mode. The dTOF sensing module is flexibly applicable to different detection scenarios.

Abstract: A processor IC has multiple power base units (PBUs), arranged in an array. Each PBU includes a switch, a memory unit and a compute unit. It further includes a power estimator for the switch, memory unit, and compute unit. The PBUs communicate with an array-level power accumulator via dedicated wiring. Communication via the dedicated wiring may use timestamps to ensure time-accurate aggregate estimates. The switch power estimator estimates dissipation based on port activity. The memory power estimator estimates dissipation based on read and write activity and bit toggling in the memory. The compute power estimator estimates power based on monitoring input data zero values, bit toggling, instruction type, and activity of reconfigurable processing units. The array-level power accumulator calculates the array-level nominal dynamic power estimate. A power clock management controller scales the dynamic power estimate for the actual clock frequency and measured supply voltage, and adds a static power estimate.

Abstract: The present invention provides a hypoosmotic stabilizer for genomic DNA of gut microbiota, comprising: 0.2% Tween 20, 0.1×PBS, and 100-200 mM guanidine thiocyanate. The hypoosmotic stabilizer for genomic DNA of gut microbiota has a pH between 7.0 and 7.4. The hypoosmotic stabilizer for genomic DNA of gut, microbiota as described in the present invention allows a gut microbiota sample to be conveniently and safely stored for 3 weeks (at room temperature) without using a refrigerator for a low storage temperature, and for even longer at a lower temperature. Moreover, the agent is non-toxic, easy-to-use, cheap, and easier to be widely used. The invention further provides a preparation method and use of the hypoosmotic stabilizer for genomic DNA of gut microbiota.

Abstract: This application provides a data processing method and an apparatus. The data processing method includes: acquiring a first frame of time-of-flight (TOF) data; processing the first frame of TOF data into a TOF image; and performing face recognition by using the first frame of TOF image to obtain a recognition result. When TOF data is used for face recognition, higher security can be a.chieved. In addition, the first frame of TOF image is used for face recognition immediately after the first frame of TOF image is obtained, so that the execution speed is high. Because the TOF data is less affected by light, a high-quality TOF image can be obtained even if the first frame of TOF data is used, and therefore, the recognition accuracy is high.

Abstract: A first frame of time of flight (TOF) data including projection off data and infrared data is obtained, and after determining that a data block satisfying that a number of data points with values greater than a first threshold is greater than a second threshold is present in the infrared data, TOF data for generating a first frame of a TOF image is obtained based on a difference between the infrared data and the projection off data. Because the data block satisfying the number of data points with values greater than the first threshold is greater than the second threshold is an overexposed data block, and the projection off data is TOF data acquired by a TOF camera with a TOF light source being off, the difference between the infrared data and the projection off data can correct the overexposure, improving quality of the first frame of the TOF image.

Abstract: A method and electronic device for implementing refocusing, and relate to the field of image processing technologies, so as to reduce costs of implementing refocusing. The method provided in the embodiments of the present invention includes acquiring at least two images that are of a to-be-photographed object and have different depths of field; acquiring an instruction operation that is performed by a user on a partial area of one of the at least two images; determining a target area according to the instruction operation; acquiring definition of the target area in the at least two images; and generating a target image according to the definition of the target area in the at least two images.