Abstract: Systems and methods for deep neural network (DNN)-based cross component prediction are provided. A method includes inputting a reconstructed luma block of an image or video into a DNN; and predicting, by the DNN, a reconstructed chroma block of the image or video based on the reconstructed luma block that is input. Luma and chroma reference information and side information may also be input into the DNN to predict the reconstructed chroma block. The various inputs may also be generated using processes such as downsampling and transformation.

Abstract: Some aspects of the disclosure provide a method of deep learning-based filtering. The method includes obtaining a reconstructed image corresponding to a coded image and at least a quantization parameter. At least a portion of the reconstructed image is reconstructed based on the quantization parameter. The method also includes generating an input to a deep learning filter according to a point multiplication operation of the reconstructed image and the at least the quantization parameter. The point multiplication operation includes pixel-wise multiplications of the at least the portion of the reconstructed image with the quantization parameter. Further, the method includes generating, by the deep learning filter and in response to the input, a filtered image corresponding to the reconstructed image. Apparatus and non-transitory computer-readable storage medium counterpart embodiments are also contemplated.

Abstract: Embodiments of this application relate to the video coding and compression field, and disclose an encoding method and apparatus, and a decoding method and apparatus, to resolve a problem that an existing split mode cannot satisfy a relatively complex texture requirement. The decoding method specifically includes: parsing a bitstream to determine a basic split mode for a current to-be-decoded picture block and a target derivation mode for a subpicture block of the current to-be-decoded picture block; splitting the current to-be-decoded picture block into N subpicture blocks in the basic split mode, where N is an integer greater than or equal to 2; deriving one derived picture block from at least two adjacent subpicture blocks in the N subpicture blocks in the target derivation mode; and decoding the derived picture block.

Abstract: A vehicle window assembly and a vehicle are provided in the disclosure. The vehicle window assembly includes multiple light guide strips spaced apart from each other, a vehicle window glass sheet, and a sealing box. The multiple light guide strips are arranged on one side surface of the vehicle window glass sheet, the sealing box is disposed at one side of each of the multiple light guide strips away from the vehicle window glass sheet, the sealing box defines grooves at positions respectively corresponding to the multiple light guide strips, and each of the multiple light guide strips is received in one of the grooves. The multiple light guide strips are received in the sealing box, the sealing box is directly fixed to the vehicle window glass sheet by the fixing member.

Abstract: A picture filtering method includes determining encoding information of a picture, the encoding information comprising classification information indicating one of an intra encoding mode or an inter encoding mode of the picture. The method further includes inputting the picture and the classification information indicating one of the intra encoding mode or the inter encoding mode into a general filtering model trained using deep learning, and obtaining a filtered picture based on the filtering model performing filtering on the picture based on the encoding information.

Abstract: A processing method includes generating first training data through a kth updated video coding application, the first training data including a sample original video frame as a training label, and a first sample to-be-filtered reconstructed frame output from the kth updated video coding application and corresponding to the sample original video frame. The method further includes, based on the sample original video frame and the first sample to-be-filtered reconstructed frame, training a second filtering model in the kth updated video coding application to obtain the second filtering model in a training convergence state, and integrating the second filtering model in the training convergence state into the kth updated video coding application to obtain a (k+1)th updated video coding application.

Abstract: The present disclosure relates to a control method of a virtual object, a control apparatus, a device and a medium. The method displays a target virtual scene in a target interactive interface, under the condition that there is a target virtual object in the plurality of virtual objects, displays a target control, the target virtual object being a virtual object controllable by the virtual subject in a non-contact manner; in response to receiving a trigger operation on the target control, causes the virtual subject to perform non-contact control on the target virtual object according to a control mode corresponding to the target control, wherein the target virtual scene includes a plurality of virtual objects and a virtual subject controlled by a user.

Abstract: Systems and methods for deep neural network (DNN)-based cross component prediction are provided. A method includes inputting a reconstructed luma block of an image or video into a DNN; and predicting, by the DNN, a reconstructed chroma block of the image or video based on the reconstructed luma block that is input. Luma and chroma reference information and side information may also be input into the DNN to predict the reconstructed chroma block. The various inputs may also be generated using processes such as downsampling and transformation.

Abstract: A deep learning-based filtering method includes obtaining a reconstructed image of luma component corresponding to an encoded image and chroma component information corresponding to the encoded image; generating an input parameter of a deep learning filter according to the reconstructed image of luma component and the chroma component information; and generating, based on the input parameter to the deep learning filter, a filtered image corresponding to the reconstructed image of luma component.

Abstract: The present disclosure relates to the technical field of medical supplies, in particular to a fusion cage. The fusion cage comprises a main body, which is a flat-shaped porous structural body and composed of a plurality of structural units. The fusion cage provided by the present disclosure is the flat-shaped porous structural body composed of the plurality of structural units, the porosity of the whole structure of the fusion cage is controlled by using the parametrization design, so the elastic modulus of a fusion body is effectively reduced, and then each structural unit comprises the basal body and the plurality of extension portions extending from the surface of the basal body, the surfaces of the basal body and/or the extension portions are composed of the plurality of curved surfaces through smooth connection, and such design effectively reduces the problem of stress concentration, and improving the postoperative recovery effect of patients.

Abstract: A video encoding method includes: acquiring a coding block of a video image frame; determining a partitioning mode available to the coding block according to a size of the coding block and an allowable maximum sub-block size ratio in response to a determination that sub-block transform (SBT) is to be performed on the coding block; determining, based on the partitioning mode available to the coding block, a target partitioning mode used during the SBT performed on the coding block; and partitioning the coding block based on the target partitioning mode, and decoding a plurality of sub-blocks obtained through the partitioning.

Abstract: A video encoding method includes: determining current coefficients to-be-decoded in a coding block of a video image frame; obtaining a flag combination corresponding to the current coefficients, the flag combination including at least two coefficient absolute level threshold flags; decoding the coefficient absolute level threshold flags in the flag combination to obtain values of the coefficient absolute level threshold flags; and determining the coefficient absolute levels of the current coefficients according to the values of the coefficient absolute level threshold flags.

Abstract: Provided are a video encoding method and apparatus, a video decoding method and apparatus, a computer-readable medium, and an electronic device, The video decoding method includes: decoding a coded block of a video image frame to obtain a target intra prediction mode corresponding to the coded block (510); in response to the target intra prediction mode belonging to a set of specified intra prediction modes, updating statistical values of the set of specified intra prediction modes, a number of the set of specified intra prediction modes being less than a total number of intra prediction modes (520); updating a most probable intra prediction mode list according to the updated statistical values of the set of specified intra prediction modes (530); and decoding another coded block of the video image frame based on the updated most probable intra prediction mode list (540).

Abstract: A video decoding method is provided. In the method, a coding block of a video image frame and a derived tree adopted by the coding block are acquired. A plurality of sub-blocks in the coding block is decoded, according to a target division mode corresponding to the derived tree, to obtain a plurality of sub-coefficient blocks. The target division mode is one of a plurality of additional division modes corresponding to the derived tree. The plurality of additional division modes is configured to divide a designated prediction block in the coding block with a side length that is not an integer power of 2 into sub-blocks with side lengths that are integer powers of 2. Reconstructed images are generated according to the derived tree adopted by the coding block and the plurality of sub-coefficient blocks.

Abstract: A video decoding method includes obtaining a size of a coding block of a video image frame, and decoding a bitstream of the coding block to obtain a value of a range variable associated with a SRCC scan region of the coding block. The range variable includes a width range variable and a height range variable. Each of the width range variable and the height range variable are obtained by comparing a prefix value to one or more predefined value intervals. The method further includes determining a horizontal coordinate of boundary coordinates of the SRCC scan region according to a difference between the width of the coding block and the width range variable and determining a vertical coordinate of the boundary coordinates of the SRCC scan region according to a difference between the height of the coding block and the height range variable.

Abstract: A data prefetching method and a terminal device are provided. The CPU core cluster is configured to deliver a data access request to a first cache of the at least one level of cache, where the data access request carries a first address, and the first address is an address of data that the CPU core cluster currently needs to access in the memory. The prefetcher in the terminal device provided in embodiments of this application may generate a prefetch-from address, and load data corresponding to the generated prefetch-from address to the first cache. When needing to access the data, the CPU core cluster can read from the first cache, without a need to read from the memory. This helps increase an operating rate of the CPU core cluster.

Abstract: A video encoding method includes: decoding a target coding unit to obtain a quantization coefficient matrix corresponding to the target coding unit; determining first reference information according to a quantization coefficient in the quantization coefficient matrix; obtaining a value of a target flag bit corresponding to the first reference information, the target flag bit being a flag bit of a target sub-block position, the target sub-block position being a position of a sub-block that requires processing of residual data in a coding unit; and determining the target sub-block position of the target coding unit according to the value of the target flag bit. A flag bit of a sub-block transform position or a transform skip sub-block position in a target coding unit is implicitly indicated using a quantization coefficient in a quantization coefficient matrix corresponding to a coding unit.

Abstract: A video decoding method includes: performing entropy decoding on a coding block of a video image frame to obtain a quantization coefficient block of residual data corresponding to the coding block; inversely quantizing the quantization coefficient block to obtain an inverse quantization coefficient matrix; inversely rearranging the inverse quantization coefficient matrix to obtain a processed coefficient matrix, non-zero coefficients in the processed coefficient matrix being concentrated in left, upper and upper left regions of the coefficient matrix; and generating the residual data according to the processed coefficient matrix.

Abstract: A video decoding method includes: performing entropy decoding processing on an encoding block of a video image frame to obtain a quantized coefficient block of residual data corresponding to the encoding block; calculating quantization coefficients in the quantized coefficient block to obtain an implicitly derived index value; determining a transformation mode of the encoding block according to the implicitly derived index value and a value of an index identifier included in the encoding block; and performing inverse transformation processing on an inverse quantization result of the quantized coefficient block based on the transformation mode of the encoding block.

Abstract: A video decoding method includes performing entropy decoding processing on a code block of a video image frame of a video image frame sequence, to obtain a quantization coefficient block of residual data corresponding to the code block; collecting statistics of quantization coefficients in a target region in the quantization coefficient block, to obtain a quantization coefficient statistical value; selecting a transform matrix combination corresponding to the quantization coefficient statistical value; and performing inverse transform processing on an inverse quantization result of the quantization coefficient block based on the selected transform matrix combination, to obtain reconstructed residual data, the reconstructed residual data being used for video image reconstruction.