Abstract: Devices, systems and methods for encoding and decoding digital video using a look-up tables (LUTs) containing motion candidates are described. In a representative aspect, a method of video processing includes maintaining one or more tables, wherein at least one table of the one or more tables is associated with one or more entries, deriving, based on a first set of motion candidates in the at least one table, a second set of motion candidates, and performing, based on the second set of motion candidates, a conversion between a current block of a video and a bitstream representation of the video.

Abstract: A secondary-side protection and sense circuit for a power converter has a sensing component, an adder amplifying circuit, an electronic switch, and a charge/discharge circuit. The sensing component is connected to an output connecting terminal of the power converter. The adder amplifying circuit has an operational amplifier, a first resistor, and a second resistor. The operational amplifier has an input terminal connected to the sensing component, an output terminal connected to a primary-side control component, and a power terminal. The first resistor and the second resistor are connected in series and between the input terminal and the power terminal of the operational amplifier. The electronic switch is connected between a ground terminal and a connection node between the first resistor and the second resistor. The charge/discharge circuit is connected to the electronic switch and the power terminal of the operational amplifier.

Abstract: The present disclosure provides a technical solution related to intelligent knowledge learning and question-answering. The technical solution involves extracting knowledge from corpus and providing a user with a question-answering service based on the extracted knowledge.

Abstract: This application provides a method of generating a description for a picture set performed at a computer device, and a storage medium. The method includes: acquiring a picture set to be processed; performing picture feature extraction on each picture in the picture set to acquire a picture feature corresponding to the picture, and forming a picture feature sequence corresponding to the picture set by using the picture features corresponding to the pictures; performing scene feature extraction on a picture feature sub-sequence corresponding to each scene in the picture feature sequence to acquire a scene feature corresponding to the scene, and forming a scene feature sequence corresponding to the picture set by using the scene features corresponding to the scenes; and generating textual description information of the picture set according to the picture feature sequence and the scene feature sequence that correspond to the picture set.

Abstract: A method for machine learning-based classification may include training a machine learning model with a full training data set, the full training data set comprising a plurality of data points, to generate a first model state of the machine learning model, generating respective embeddings for the data points in the full training data set with the first model state of the machine learning model, applying a clustering algorithm to the respective embeddings to generate one or more clusters of the embeddings, identifying outlier embeddings from the one or more clusters of the embeddings, generating a reduced training data set comprising the full training data set less the data points associated with the outlier embeddings, training the machine learning model with the reduced training data set to a second model state, and applying the second model state to one or more data sets to classify the one or more data sets.

Abstract: Disclosed is a dual-input circuit, adopting a contactor as a switching device, and being more suitable for medium and high-power applications. In addition, the adopted contactor is a single double-pole double-throw contactor with an interlocking function. In this way, the circuit is safe, simple, and easily controllable, and space utilization can be further improved. In addition, the adopted contactor further is of a magnetic attraction type, with lower power consumption for long-term work. In the present invention, either one of two inputs is set as a main input and the other is set to a backup input by default, so that the present invention can be more flexibly applied and is more applicable. To enable switching when both or only either one of the two inputs has an input voltage, the present invention further adopts a single dual-input auxiliary power supply to reduce a required space and further reduce costs.

Abstract: Disclosed is a submarine network device, comprising a fiber set, a pump laser set, an erbium doped fiber amplifier (EDFA) set, a primary fiber coupler (CPL) set and a secondary CPL set, wherein the primary CPL set comprises N primary CPLs, the secondary CPL set comprises N secondary CPLs, with N being an integer greater than or equal to 3. The fiber set is configured to connect the pump laser set, the primary CPL set, the secondary CPL set and the EDFA set. An input port of each primary CPL in the primary CPL set is at least connected with a pump laser. An output port of each secondary CPL in the secondary CPL set is at least connected with an EDFA. Output ports of each primary CPL in the primary CPL set are respectively connected with two different secondary CPLs that are spaced by a secondary CPL, and input ports of each secondary CPL in the secondary CPL set are respectively connected with two different primary CPLs that are spaced by a primary CPL.

Abstract: An optical phased array, a method for reducing a phase error thereof, a LiDAR, and an intelligent apparatus are provided. The optical phased array includes an optical signal output unit, a waveguide unit, and an antenna transmitting unit. The optical signal output unit is configured to output M optical signals. The waveguide unit includes M waveguide pipes, each waveguide pipe includes at least one connection waveguide, and each of the at least one connection waveguide includes an input mode converter, a wide waveguide, and an output mode converter that are connected in sequence. The antenna transmitting unit is configured to transmit M optical signals outputted from the waveguide unit.

Abstract: There is disclosed a waveguide for use in an augmented reality or virtual reality display. The waveguide comprises a plurality of optical structures in a photonic crystal. The plurality of optical structures are arranged in an array to provide at least two diffractive optical elements. Each of the two diffractive optical elements is configured to receive light from an input direction and couple it towards the other diffractive optical element which can then act as an output diffractive optical element, providing outcoupled orders towards a viewer. The plurality of optical structures respectively have a shape, when viewed in the plane of the waveguide, comprising twelve substantially straight sides, six of the sides having respective normal vectors at a first angle, and the other six of the sides having respective normal vectors at a second angle which is different to the first angle.

Abstract: A data labeling method and device and a computer-readable storage medium. The method includes: based on a knowledge graph, remotely acquiring data to be labeled; performing data cleaning and pre-labeling to the data to be labeled, to obtain pre-labeled data; and performing labeling correction to the pre-labeled data.

Abstract: Disclosed are a pallet configured to bear a battery cell. The pallet includes an external fixing rack and a plurality of supporting beams. A mounting space is formed inside the external fixing rack. The plurality of supporting beams are disposed in the mounting space. The plurality of supporting beams are arranged along a first direction. The plurality of supporting beams are connected to the external fixing rack separately. At least two of the supporting beams are spaced apart. The plurality of supporting beams are configured to bear the battery cell.

Abstract: A semiconductor package includes a package substrate; semiconductor devices disposed on the package substrate; a package ring disposed on a perimeter of the package substrate surrounding the semiconductor devices; a cover including silicon bonded to the package ring and covering the semiconductor devices; and a thermal interface structure (TIS) thermally connecting the semiconductor devices to the cover.

Abstract: Embodiments for video encoding and video decoding are described. One example method includes performing a conversion between a video and a bitstream of the video, wherein the bitstream includes one or more pictures including one or more subpictures according to a rule, and wherein the rule specifies that, responsive to a condition, a rewriting operation is performed on referenced one or more sequence parameter sets during a subpicture sub-bitstream extraction process by which a target output sub-bitstream is extracted from the bitstream.

Abstract: A pancreatin preparation having reduced viral infectivity includes one or more pancreatin enzymes and peracetic acid (PAA). At least one pancreatin enzyme may be derived from an animal source such as a porcine pancreas gland. In a particular embodiment, the one or more pancreatin enzymes are selected from the group consisting of lipases, proteases, and amylases.

Abstract: Provided herein are compounds useful in the treatment of inflammatory diseases, pharmaceutical compositions comprising the same, and methods of use and preparation thereof. The compounds exhibit inhibitory effects on the expression and secretion of pro-inflammatory cytokines, such as IL-1?, IL-6 and TNF-?.

Abstract: In an example embodiment, a solution is used to provide container volume replication via a container storage replication log and volume buffer synchronization, which is built on top of a container cloud platform whose container metadata and replication runtime configuration are all managed by a storage manager (a service orchestrated by its job scheduler and service orchestrator). This container volume replication ensures the data security for a long-running service in the container. In the case of any disaster, the in-memory database and application data inside of the container can be recovered via volume replication. This provides container volume replication for long-running containerized applications whose states keep changing.

Abstract: A light guide plate assembly, a backlight module, and a display device are provided. The light guide plate assembly is configured to guide exiting light emitted from a light-emitting element. The light guide plate assembly includes a reflective part includes an opening, and a transmissive part. At least a part of the transmissive part is located at a side of the reflective part adjacent to a light-exiting side of the light guide plate assembly. The transmissive part includes a first part. Along a direction from a center of the opening to an edge of the opening, light intensity of the exiting light received by the first part is negatively correlated to light transmittance of the first part.

Abstract: A signal processing chip includes: a receiving module, configured to receive a WLAN analog baseband signal from a baseband chip; an analog-to-digital conversion module, configured to convert the WLAN analog baseband signal into a WLAN digital baseband signal; a processing module, configured to process the WLAN digital baseband signal into a WLAN analog intermediate frequency signal; and a sending module, configured to send the WLAN analog intermediate frequency signal to a radio frequency processing apparatus.

Abstract: A pixel arrangement structure, a display method and a preparing method of a pixel arrangement structure, and a display substrate are provided. The pixel arrangement structure includes first color sub-pixel blocks, second color sub-pixel blocks, and third color sub-pixel blocks. Each of the plurality of minimum repeating regions has a rectangular shape and includes a first virtual rectangle, and the first virtual rectangle includes one first color sub-pixel block, one second color sub-pixel block and one third color sub-pixel block. Any edge of the first virtual rectangle has a non-zero included angle with a first direction, and the first direction is a row direction or a column direction. The first color sub-pixel block is on a perpendicular bisector of the first edge, the second color sub-pixel block and the third color sub-pixel block are on different sides of the perpendicular bisector of the first edge.