Abstract: A method, computer program, and computer system is provided for coding a video sequence. Video data comprising one or more luma samples is received. A YUV format associated with the received video data is determined. A height setting value and a width setting value corresponding to the one or more luma samples are restricted based on the determined YUV format. The YUV format may include a YUV4:4:4 format, a YUV4:2:2 format, a YUV4:2:0 format, and a YUV4:0:0 format.

Abstract: Systems and methods for pre-training a dialogue model with semantic information include: generating a dialogue-level abstract meaning representation (AMR) graph for an input dialogue associated with a speaker, learning core semantic units of the input dialogue based on nodes of the dialogue-level AMR graph, learning semantic relations between words of a sentence of the input dialogue based on edges of the dialogue-level AMR graph, learning an overall agreement of the input dialogue and the dialogue-level AMR graph, and training the dialogue model based on the learned core semantic units, semantic relations between words, and overall agreement.

Abstract: Systems and methods for training a model to perform end-to-end character-to-phoneme (C2P) conversion include: selecting a plurality of unlabeled sentences from a first data source, selecting a plurality of labeled sentences from a second data source, preprocessing a combined corpus of the selected unlabeled and labeled sentences to extract a plurality of linguistic features, generating mixed training data by automatically labeling tokens in the preprocessed corpus based on the plurality of extracted linguistic features, and training a pre-trained model, using the mixed training data, to perform end-to-end C2P conversion.

Abstract: In a method, a visual hull is generated based on intersections of first 3D primitives of a plurality of first silhouettes with a bounding box of a 3D model. The first silhouettes are generated by projecting the 3D model onto planes perpendicular to a number of selected view directions of the 3D model. Each of the first 3D primitives is obtained by extruding a connected loop of a respective first silhouette along a view direction of the number of selected view directions that is associated with the respective first silhouette. A carved mesh is be generated based on subtractions of second 3D primitives derived from positive parts of the 3D model to carve out redundant structures from the visual hull. The positive parts are obtained based on fitting planes that slices the 3D model. A low-poly mesh sequence is generated based on progressive simplifications of the carved mesh.

Abstract: A method, computer program, and computer system for encoding or decoding video data, and indicating, with a syntax element, types of slices for all slices of a coded picture, the syntax element being coded using an unsigned integer.

Abstract: A method of acoustic echo suppression using a recurrent neural network, performed by at least one processor, is provided. The method includes receiving a microphone signal and a far-end reference signal, estimating an echo suppressed signal and an echo signal based on the microphone signal and the far-end reference signal, estimating enhancement filters for the microphone signal based on the echo suppressed signal and the echo signal, generating an enhanced signal based on the enhancement filters, and adjusting the enhanced signal using automatic gain control (AGC) and outputting the adjusted signal.

Abstract: A method includes receiving video data including one or more subpicture. The video data includes a flag, signaled in a sequence parameter set, indicating whether to use a network abstraction unit (NAL) layer identifier as a subpicture identifier, where the subpicture identifier is signaled in a NAL unit header based on the flag indicating to use the NAL layer identifier, and the subpicture identifier is signaled in a slice header based on the flag indicating not to use the NAL layer identifier. The video data is decoded based on the subpicture identifier.

Abstract: A method of and an apparatus for controlling intra prediction for decoding of a video sequence are provided. The method may include based on a reference line index signaling a first reference line nearest to a coding unit, among a plurality of reference lines adjacent to the coding unit, applying intra smoothing on one or more first reference lines comprising the first reference line, among the plurality of reference lines, and based on the intra smoothing being applied on the one or more first reference lines, applying a position-dependent intra prediction combination (PDPC) on one or more third reference lines comprising the first reference line, among the plurality of reference lines, while preventing application of the PDPC on one or more fourth reference lines other than the one or more third reference lines, among the plurality of reference lines.

Abstract: A method of video decoding performed in a video decoder is disclosed. A syntax element can be received from a bitstream of a coded video indicating that residual blocks of a current coding unit (CU) are processed with a color space conversion. The residual blocks of the current CU can include a luma residual block, a first chroma residual block, and a second chroma residual block. In response to receiving the syntax element indicating that the residual blocks of the current CU are processed with the color space conversion, one of two options of color space conversion equations can be selected based on a transform skip flag and a quantization parameter (QP) corresponding to each of the residual blocks of the current CU. An inverse color space conversion can be applied using the selected option of color space conversion equations to the residual blocks of the current CU.

Abstract: Systems, devices, and methods for reporting user equipment activity during a 5G Media Streaming (5GMS) uplink streaming session are provided, which may include initializing contribution reporting based on selected user preferences during media streaming during the 5GMS uplink streaming session, and transmitting a contribution report associated with the media streaming during the 5GMS uplink streaming session. The approach may further include generating a final contribution report associated with the media streaming during the 5GMS uplink streaming session in response to receiving a signal to stop media streaming.

Abstract: An image processing method includes obtaining statistics information of a block of an image frame, determining whether the statistics information satisfies a condition, and disabling cross-component filtering (CCF) for the block in response to the statistics information satisfying the condition.

Abstract: A method of video encoding includes determining whether a plurality of constraint flags of a constraint information group are to be signaled for a subset of coding blocks, and signaling a gate flag indicating whether the plurality of constraint flags of the constraint information group are signaled. The gate flag is signaled in constraint information in a syntax structure of a coded video bitstream. In accordance with the signaled gate flag indicating that the plurality of constraint flags of the constraint information group are signaled, the method further includes signaling the plurality of constraint flags of the constraint information group in the constraint information in the syntax structure of the coded video bitstream. In accordance with the plurality of constraint flags being signaled in the constraint information, the method further includes encoding the subset of coding blocks based on the plurality of constraint flags.

Abstract: A method of managing capabilities of a media streaming network using at least one processor, including receiving a capability request for media streaming capabilities of an edge data network (EDN); determining the media streaming capabilities of the EDN; transmitting a capability response based on the determined media streaming capabilities; receiving a media processing workflow request based on the capability response; establishing a media streaming session according to the media processing workflow request; and streaming media content based on the media streaming session.

Abstract: A method of substitutional neural residual compression is performed by at least one processor and includes estimating motion vectors, based on a current image frame and a previous reconstructed image frame, obtaining a predicted image frame, based on the estimated motion vectors and the previous reconstructed image frame, and subtracting the obtained predicted image frame from the current image frame to obtain a substitutional residual. The method further includes encoding the obtained substitutional residual, using a first neural network, to obtain an encoded representation, and compressing the encoded representation.

Abstract: Aspects of the disclosure include methods, apparatuses, and non-transitory computer-readable storage mediums for video encoding/decoding. An apparatus includes processing circuitry that receives metadata associated with a coded video bitstream. The metadata includes labeling information of one or more objects detected in a first picture that is coded in the coded video bitstream. The processing circuitry decodes the labeling information of the one or more objects in the first picture that is coded in the coded video bitstream. The processing circuitry applies the labeling information to the one or more objects in the first picture.

Abstract: Whether each of a plurality of duplicate vertices is skippable in a mesh is determined. One or more skippable duplicate vertices of the plurality of duplicate vertices in the mesh are excluded to generate an updated mesh. Vertices of the updated mesh are divided into a plurality of vertex groups, where the plurality of vertex groups includes a first vertex group. A prediction mode of the first vertex group is determined at least based on estimation errors of vertices in the first vertex group of the updated mesh. Each of the estimation errors indicates a difference between a respective vertex in the first vertex group and a corresponding reference vertex in the reference frame. Prediction information of the vertices in the first vertex group is generated based on the determined prediction mode.

Abstract: In a method of video encoding for an encoder, a determination is made to determin whether (i) an implicit transform scheme is enabled, and (ii) at least one of a low-frequency non-separable transform (LFNST) and a matrix-based intra predication mode (MIP) is invalid for a coding unit (CU). In response to the determination that the implicit transform scheme is enabled, and at least one of the LFNST and MIP is invalid, a primary transform type is determined based on a size of the CU. A primary transform is performed for a transform block that is partitioned from the CU in accordance with the determined primary transform type. A coded bitstream that indicates the primary transform type of the CU is subsequently ouput.

Abstract: A method of and an apparatus for controlling intra prediction for encoding of a video sequence are provided. The method includes determining a ratio of a width to a height of a coding unit, and based on the determined ratio being different than one, adding, to a table including intra prediction modes corresponding to intra prediction angles, first wide angles toward a bottom-left edge of the coding unit, second wide angles toward a top-right edge of the coding unit, and additional intra prediction modes respectively corresponding to the first wide angles and the second wide angles. The method further includes signaling, for encoding the video sequence, one of the intra prediction modes and the additional intra prediction modes added to the table.

Abstract: A method of decoding an encoded video bitstream using at least one processor, including: obtaining from the encoded video bitstream a coded video sequence including a picture unit corresponding to a coded picture; obtaining a picture header (PH) network abstraction layer (NAL) unit included in the picture unit; obtaining at least one video coding layer (VCL) network abstraction layer (NAL) unit included in the picture unit; decoding the coded picture based on the PH NAL unit, the at least one VCL NAL unit, and an adaptation parameter set (APS) included in an APS NAL unit obtained from the coded video sequence; and outputting the decoded picture, wherein the APS NAL unit is available to the at least one processor before the at least one VCL NAL unit.

Abstract: Systems and methods for encoding and decoding using warp motion modes are provided. A method includes receiving a bitstream that includes a coded picture; obtaining warp information that is signaled in the bitstream, the warp information including a warp type indicating a number of parameters to be corrected by delta values, the number of parameters defining warp motion for predicting a current block of the coded picture; obtaining at least one delta value from the bitstream that is a delta of at least one parameter that defines the warp motion for predicting the current block; obtaining a warp model based on the warp information and the at least one delta value; and predicting the current block by performing warped motion compensation using the warp model.