Abstract: The present invention provides a method for decoding a video signal including extracting an inter prediction mode of a current block from the video signal; generating two non-square prediction blocks using two motion vectors according to the inter prediction mode; performing a non-square transform on the two non-square prediction blocks to obtain two non-square prediction blocks on a frequency domain; updating the two non-square prediction blocks on the frequency domain using a correlation coefficient or a scaling coefficient; and generating a reconstruction block based on the updated non-square prediction blocks on the frequency domain and a residual block.

Abstract: Aspects of the disclosure provide methods and apparatuses for point cloud compression and decompression. In some examples, an apparatus for point cloud compression/decompression includes processing circuitry. For example, the processing circuitry of the apparatus for point cloud encoding determines, from a point cloud, more than two candidate source points that are associated with a target point in a reconstructed geometry reconstructed from a compressed geometry image for the point cloud. Then the processing circuitry determines a color for the target point based on colors of the more than two candidate source points, and encodes texture of the point cloud with the target point having the determined color.

Abstract: Aspects of the disclosure provide methods and apparatuses for point cloud compression and decompression. In some examples, an apparatus for point cloud compression/decompression includes processing circuitry. For example, the processing circuitry determines, from a point cloud, at least a first set of candidate source points that are associated with a target point in a geometry image reconstructed from a compressed geometry image for the point cloud. Then, the processing circuitry weights a color of each candidate source point in the first set based on a combination of a location difference between the candidate source point and the target point, and a color difference between the color of the candidate source point and an aggregate color of the first set of candidate source points. Further, the processing circuitry determines a color of the target point based on a first weighted color average of the first set of candidate source points.

Abstract: A method of dynamic point cloud partition packing is by at least one processor and includes obtaining one or more region of interest (ROI) patches from an ROI of a point cloud, and attempting to pack, into one among tiles of a tile map, one among the obtained one or more ROI patches, in a tile scan order. The method further includes identifying whether the one among the one or more ROI patches is packed successfully into the one among the tiles, and based on the one among the one or more ROI patches being determined to be not packed successfully into the one among the tiles, chunking the one among the one or more ROI patches into multiple ROI patches.

Abstract: The present invention provides a method for encoding a video signal on the basis of a graph-based separable transform (GBST), the method comprising the steps of: generating an incidence matrix representing a line graph; training a sample covariance matrix for rows and columns from the rows and columns of a residual signal; calculating a graph Laplacian matrix for rows and columns on the basis of the incidence matrix and the sample covariance matrix for rows and columns; and obtaining a GBST by performing eigen decomposition of the graph Laplacian matrix for rows and columns.

Abstract: The present invention provides a method for encoding a video signal based on an Edge Adaptive Graph-Based Transform (EA-GBT) including detecting a step edge or a ramp edge from a residual signal; generating a graph signal based on at least one of the step edge or the ramp edge; obtaining an EA-GBT coefficient by performing the EA-GBT for the graph signal; quantizing the EA-GBT coefficient; and entropy-encoding the quantized EA-GBT coefficient.

Abstract: A method, apparatus, and computer-readable medium for point cloud coefficient coding are provided. Transform coefficients associated with point cloud data are decomposed into set-index values and symbol-index values, the symbol index-value specifying location of the transform coefficient within a set. The decomposed transform coefficients are partitioned into one or more sets based on the set-index values and the symbol-index values. The set-index values of the partitioned transform coefficients are entropy-coded, and the symbol-index values of the partitioned transform coefficients are bypass-coded. The point cloud data is compressed based on the entropy-coded symbol-index values and the bypass-coded set-index values.

Abstract: A method, computer system, and computer-readable medium are provided for point cloud attribute coding by at least one processor. Data associated with a point cloud is received. The received data is transformed through a lifting decomposition based on enabling a scalable coding of attributes associated with the lifting decomposition. The point cloud is reconstructed based on the transformed data.

Abstract: A method of point cloud attribute coding is performed by at least one processor and includes receiving a point cloud data set comprising a bounding box. A determination is made as to whether the bounding box is symmetric or asymmetric. A quad-tree-binary-tree (QtBt) decomposition is performed on the point cloud data set based on the bounding box being asymmetric.

Abstract: The present invention provides a method for performing a transform, the method comprising the steps of: deriving a row transform set, a column transform set, and a permutation matrix on the basis of a given transform matrix (H) and error tolerance parameter; obtaining a row-column transform (RCT) coefficient on the basis of the row transform set, the column transform set, and the permutation matrix; and performing a quantization and an entropy encoding on the RCT coefficient, wherein the permutation matrix represents a matrix obtained by permutating a row of an identity matrix.

Abstract: A method of processing point cloud data at a decoder can include receiving three dimensional (3D) coordinates of a set of points of a point cloud including first points and a current point. Each of the first points can be associated with a reconstructed attribute value. A group of neighboring points of the current point can be determined from the first points. A first index is received and the first index indicates a reconstructed attribute value selected from the reconstructed attribute values of the plurality of neighboring points. The reconstructed attributed value indicated by the first index is determined based on a rate-distortion decision function. An attribute distance for each of the neighboring points can be determined based on the reconstructed attribute values of the neighboring points.

Abstract: A method of interframe point cloud attribute coding is performed by at least one processor and includes obtaining, as a motion estimation unreliability measure of motion estimation of a target frame, a value inversely proportional to a ratio of a number of first point cloud samples of the target frame respectively with second point cloud samples of an interframe reference frame, to a number of point cloud samples in the target frame. The method further includes identifying whether the obtained motion estimation unreliability measure is greater than a predetermined threshold, based on the obtained motion estimation unreliability measure being identified to be greater than the predetermined threshold, skipping motion compensation of the target frame, and based on the obtained motion estimation unreliability measure being identified to be less than or equal to the predetermined threshold, performing the motion compensation of the target frame.

Abstract: A method of point cloud attribute coding performed by at least one processor, including obtaining an encoded bitstream corresponding to a point cloud; determining whether the encoded bitstream was encoded using an inter-channel tool for inter-channel decorrelation; based on determining that the encoded bitstream was encoded using the inter-channel tool, decoding the encoded bitstream using the inter-channel tool to reconstruct an attribute signal corresponding to the point cloud; and reconstructing the point cloud using the reconstructed attribute signal.

Abstract: A method of point cloud attribute coding includes calculating a motion-estimating uncertainty value based on geometry data associated with a point cloud; identifying one or more inter-frame nearest neighbor point cloud samples corresponding to the point cloud in response to a determination that the motion-estimating uncertainty value is less than a threshold value; ranking one or more temporal candidate points associated with the one or more identified inter-frame nearest neighbor point cloud samples based on the motion-estimation uncertainty value; and extracting one or more sample attribute values from the one or more temporal candidate points, whereby the one or more sample attribute values correspond to the geometry data.

Abstract: A method of point cloud attribute coding includes obtaining an attribute signal corresponding to a point cloud; determining whether lossless lifting is enabled; based on determining that lossless lifting is enabled, modifying at least one from among a plurality of quantization weight and a plurality of lifting coefficients; decomposing the attribute signal into a plurality of detail signals and a plurality of approximation signals based on the modified at least one from among the plurality of quantization weights and the plurality of lifting coefficients; generating a bitstream representing the point cloud based on the plurality of detail signals and the plurality of approximation signals; and transmitting the bitstream.

Abstract: A method and device for coding attribute information of a point cloud may include obtaining the point cloud including a set of points in a three-dimensional space.

Abstract: There is included a method and apparatus comprising computer code configured to cause a processor or processors to perform obtaining a leaf node of geometry based point cloud compression (G-PCC) data, splitting the leaf node into a plurality of cuboids, deriving separate triangle soups for each of the cuboids, and coding a plurality of flags respectively for each of the edges of each of the cuboids, where the plurality of flags indicate whether vertices of the separate triangle soups are present on ones of the edges.

Abstract: A method and apparatus for referencing a parameter set to permit decoding of a coded point cloud sequence includes parsing a parameter set associated with the coded point cloud sequence. A dependent parameter set is identified based on parsing the parameter set. It is identified whether the dependent parameter set is a patch-level parameter set. The patch-level parameter set is referenced based on identifying that the dependent parameter set is the patch-level parameter set to permit the decoding of the coded point cloud sequence. A frame-level parameter set is referenced based on identifying that the dependent parameter set is not the patch-level parameter set to permit the decoding of the coded point cloud sequence.

Abstract: A method of processing point cloud data at a decoder can include receiving three dimensional (3D) coordinates of a set of points of a point cloud including first points and a current point. Each of the first points can be associated with a reconstructed attribute value. A group of neighboring points of the current point can be determined from the first points. An attribute distance for each of the neighboring points can be determined based on the reconstructed attribute values of the neighboring points. An attribute prediction of the current point can be determined based on the attribute distances of the neighboring points.

Abstract: Methods and apparatuses of decoding a video stream encoded using video point cloud coding include obtaining a geometry-reconstructed point cloud; dividing the geometry-reconstructed point cloud into a plurality of cells; obtaining a first reference point including a centroid of a first plurality of points; generating a second plurality of points by applying a first filter to the first plurality of points based on the first reference point; obtaining a second reference point including a centroid of the second plurality of points; generating a third plurality of points by applying a second filter to the second plurality of points based on the second reference point, wherein a strength of the first filter is higher than a strength of the second filter; obtaining a smoothed geometry-reconstructed point cloud based on the third plurality of points; and reconstructing a dynamic point cloud using the smoothed geometry-reconstructed point cloud.