Abstract: An apparatus for point cloud compression includes processing circuitry. The processing circuitry represents points in a point cloud using a spherical coordinate system having a first angular dimension, a second angular dimension and a radial dimension. A point is represented by a first angular coordinate in the first angular dimension, a second angular coordinate in the second angular dimension, a radial coordinate in the radial dimension and at least an attribute value. The processing circuitry forms at least a two-dimensional (2D) matrix with a first dimension corresponding to the first angular dimension and a second dimension corresponding to the second angular dimension. The point is placed as an element in the 2D matrix based on the first angular coordinate and the second angular coordinate. Further, the processing circuitry compresses data of the 2D matrix, and generates a bitstream that includes compressed data of the 2D matrix.

Abstract: A method of point cloud geometry encoding includes receiving a slice of a point cloud frame for encoding, and constructing an octree representing a geometry of points in a bounding box of the slice where a current node of the octree is partitioned with a quadtree (QT) partition or a binary tree (BT) partition. The constructing includes determining a value of a partitionSkip variable specifying a partition type and a partition direction of the current node of the octree.

Abstract: In a method of point cloud geometry decoding in a point cloud decoder, chroma prediction residual information of a point in a set of points is received from a coded bitstream for a point cloud that includes the set of points. The chroma prediction residual information includes a Cb component and a Cr component. Further, a type of correlation between the Cb component and the Cr component of the chroma prediction residual information is determined by processing circuitry and from the coded bitstream for the point cloud. The chroma prediction residual information is decoded based on the type of the correlation between the Cb component and the Cr component of the chroma prediction residual information.

Abstract: Aspects of the disclosure provide methods and apparatuses for use in video coding for machines. In some examples, an apparatus of video coding for machines includes processing circuitry. The processing circuitry determines a first picture quality vs. coding efficiency characteristic for a first coding scheme of video coding for machines (VCM), and a second picture quality vs. coding efficiency characteristic for a second coding scheme of VCM. Then, the processing circuitry determines a Bjøntegaard delta (BD) metric for a comparison of the first coding scheme and the second coding scheme, based on the first picture quality vs. coding efficiency characteristic and the second picture quality vs. coding efficiency characteristic.

Abstract: An embodiment of the present disclosure discloses an intra-prediction-based encoding and decoding method, apparatus and filter, and belongs to a video coding technical field.

Abstract: A production line for producing components to a high standard of cleanliness and sealed and protected in that state includes a loading device, a cleaning device, a detecting device, a pasting device, a heat-sealing device, a packing device, and transfer devices of the production line. The production line automatically processes the components for obtaining components with the high cleanliness. By the processes of protective film pasting, heat-sealing, and packing, the components may be further protected from subsequent pollution. A method for producing components with a high cleanliness applied to the production line is also disclosed.

Abstract: In a method of point cloud attribute decoding in a point cloud decoder, a sorting method of a set of points of a point is determined. A quantization parameter for quantizing transform coefficients of a Haar-based attribute coding of the point cloud is determined based on first signaling information in a coded bitstream of the point cloud. The Haar-based attribute coding is configured to perform point cloud compression on the set of points of the point cloud based on a Haar transform. An entropy decoding method for decoding the quantized transform coefficients is determined according to the quantization parameter that is determined based on the first signaling information. The quantized transform coefficients are further determined based on the entropy decoding method. Attributes of the point cloud are reconstructed based on the determined sorting method and the determined quantized transform coefficients.

Abstract: Aspects of the disclosure include methods, apparatuses, and non-transitory computer-readable storage mediums for point cloud compression. An apparatus includes processing circuitry that encodes information associated with a current point of a plurality of points of a point cloud. The plurality of points is partitioned into multiple bounding boxes. The processing circuitry determines whether a first size of a hash table is greater than or equal to a predetermined maximum size of the hash table. The processing circuitry removes information associated with non-boundary points in the multiple bounding boxes from the hash table based on the first size of the hash table being greater than or equal to the predetermined maximum size of the hash table. The processing circuitry stores the encoded information associated with the current point into the hash table.

Abstract: A method, computer program, and computer system is provided for video coding. Video data including one or more frames is received. A static background is estimated for each of the one or more frames based on a temporal average of the one or more frames. Pixels from among the one or more frames are identified as corresponding to the static background. Noise is removed in the static background based on the identified pixels.

Abstract: A method, computer program, and computer system is provided for decoding point cloud data. Data corresponding to a point cloud is received. Hash elements corresponding to one or more neighboring nodes associated with a current node are identified. A size of a hash table is decreased based on deleting one or more of the hash elements corresponding to non-border regions of the one or more neighboring nodes. The data corresponding to the point cloud is decoded based on the hash table having the decreased size.

Abstract: Aspects of the disclosure provide methods, apparatuses, and a non-transitory computer-readable medium for point cloud compression and decompression. In a method, an octree partitioning structure for a bounding box of a point cloud is determined. Whether parallel encoding is to be performed on occupancy codes of nodes in a range of one or more partitioning depths in the octree partitioning structure is determined. The parallel encoding on the occupancy codes of the nodes is performed in response to the parallel encoding being determined to be performed on the occupancy codes of the nodes in the range of the one or more partitioning depths in the octree partitioning structure. A coded bitstream is generated based on the encoded occupancy codes of the nodes. The coded bitstream includes syntax information that indicates whether parallel decoding is to be performed on the occupancy codes of the nodes in the range.

Abstract: A method for coding information of a point cloud comprises obtaining the point cloud including a set of points in a three-dimensional space; partitioning the point cloud into a plurality of objects and generating occupancy information for each of the plurality of objects; and encoding the occupancy information by taking into account the distance between the plurality of objects.

Abstract: Aspects of the disclosure provide methods and apparatuses for point cloud compression. In some examples, an apparatus for point cloud compression includes processing circuitry. In some embodiments, the processing circuitry determines one or more original points in a point cloud that are associated with a reconstructed position. Positions of the one or more original points can be reconstructed, according to a geometry quantization, to the reconstructed position. The processing circuitry then determines an attribute value for the reconstructed position based on attribute information of the one or more original points, and encodes texture of the point cloud with the reconstructed position having the determined attribute value.

Abstract: Provided are a point cloud attribute prediction method and prediction device based on a filter. The method comprises a coding method and a decoding method, and the device comprises a coding device and a decoding device. The method comprises: determining K nearest neighbor points of the current point; determining a filter matrix; and determining an attribute prediction value of the current point according to the filter matrix. Therefore, the compression performance of a point cloud attribute can be improved by means of selecting an appropriate filter.

Abstract: Aspects of the disclosure include a method for point cloud coding. In the method, whether decoding of occupancy codes of nodes in a range of octree partition depths in an octree partitioning structure of a point cloud reaches a minimum octree partition depth at which parallel decoding is enabled is determined. Arithmetic coding information for decoding the occupancy codes of the nodes in the minimum octree partition depth is stored based on the decoding of the occupancy codes of the nodes in the range of octree partition depths reaching the minimum octree partition depth. The parallel decoding is performed on occupancy codes of the nodes in each of the at least one remaining octree partitions depth based on the stored arithmetic coding information. The point cloud is reconstructed based on the occupancy codes of the nodes in the range of octree partition depths in the octree partitioning structure.

Abstract: A tone mapping method is provided. This method includes: obtaining one or a plurality of high dynamic range images, and determining a storage format of each high dynamic range image; performing an image decomposition on the high dynamic range image to obtain a first component, a second component and a third component, when the storage format of the high dynamic range image is determined as a predetermined storage format; inputting the first component and the second component into a predetermined deep neural network, and using the deep neural network to perform mapping on the first component and the second component respectively to obtain a first mapped component and a second mapped component; and fusing the first mapped component and the second mapped component with the third component to obtain a fused low dynamic range image corresponding to the high dynamic range image.

Abstract: Aspects of the disclosure provide methods, apparatuses, and a non-transitory computer-readable medium for point cloud coding. In a method, when parallel octree coding is enabled for occupancy codes of nodes in an octree partitioning structure of the point cloud, syntax information of the point cloud is decoded from a coded bitstream and a bitstream offset of an octree depth is determined. The syntax information indicates a bitstream length of the octree depth at which the parallel octree coding is enabled. Parallel decoding is performed on the occupancy codes of the nodes of the octree depth based on the bitstream offset and the bitstream length of the octree depth. Further, the point cloud is reconstructed based on the occupancy codes of the nodes.

Abstract: A method, a non-transitory computer readable medium, and a computer system is provided for encoding or decoding video data. The method may include: receiving an entropy coded bitstream comprising compressed video data including point cloud occupancy codes; generating one or more dequantized dimensions of a boundary box of a point cloud; based on determining that the node or node depth has attribute information, using the attribute information for the node or node depth; and based on determining that the node or node depth does not have attribute information, obtaining attribute information for the node or node depth by inheriting attribute information from a node or node depth that is at least one depth level in an octree of the point cloud above a depth level of the node or node depth in the octree.

Abstract: The present application discloses a cross-media retrieval method based on deep semantic space, which includes a feature generation stage and a semantic space learning stage. In the feature generation stage, a CNN visual feature vector and an LSTM language description vector of an image are generated by simulating a perception process of a person for the image; and topic information about a text is explored by using an LDA topic model, thus extracting an LDA text topic vector. In the semantic space learning phase, a training set image is trained to obtain a four-layer Multi-Sensory Fusion Deep Neural Network, and a training set text is trained to obtain a three-layer text semantic network, respectively. Finally, a test image and a text are respectively mapped to an isomorphic semantic space by using two networks, so as to realize cross-media retrieval. The disclosed method can significantly improve the performance of cross-media retrieval.

Abstract: In some examples, an apparatus for point cloud compression/decompression includes processing circuitry. The processing circuitry determines a flag that indicates an enable/disable control for saving coding state in a largest coding unit (LCU) based coding of a point cloud. In some examples, the processing circuitry stores coding state information before a coding of a first LCU; and in response to the flag indicating an enable control, the processing circuitry restores, a coding state according to the stored coding state information before a coding of a second LCU. In some examples, in response to the flag indicating an enable control, the processing circuitry stores the coding state information before the coding of the first LCU. In some examples, in response to the flag indicating a disable control, the processing circuitry skip the storing/restoring of the coding state information.