Abstract: A reference specifying unit specifies reference values corresponding to a plurality of points of point cloud data. A residual decoding unit decodes attribute residuals of the plurality of points from encoded residual data obtained by encoding the attribute residuals, the attribute residuals of the plurality of points being residuals of attribute values of the plurality of points with respect to the reference values. An attribute value calculation unit calculates, for each of the plurality of points, an attribute value of the point from an attribute residual and a reference value.

Abstract: According to an aspect of the present invention, there is provided a learning device including: a classification unit that classifies latent variables obtained from learning data used for learning into a label feature quantity and a non-label feature quantity; a decoding unit that decodes the label feature quantity and the non-label feature quantity classified by the classification unit by using decoder parameters to generate reconstruction data; and an optimization unit that optimizes the decoder parameters to minimize a classification error between the label feature quantity and label information used for classification by using the label feature quantity, and minimize a reconstruction error by using the label feature quantity and the non-label feature quantity.

Abstract: A decoding method is a decoding method performed by a decoding device, the decoding method including steps of: acquiring, from among a plurality of slice numbers indicating a plurality of slices associated with encoded data of the entire point cloud distributed in a three-dimensional space, a first slice number indicating a first slice associated with encoded data of a point cloud of a first space that is a portion of the three-dimensional space, on the basis of a pointer indicating the first slice; and acquiring the encoded data of the point cloud of the first space on the basis of the first slice number, and decoding point cloud data of the first space from the encoded data of the point cloud of the first space.

Abstract: According to an aspect of the present invention, there is provided a learning device including: a classification unit that classifies latent variables, which are feature quantities obtained from learning data used for learning, by using a label feature quantity having label information used for classification; a decoding unit that decodes the latent variables to generate reconstruction data by using predetermined decoding parameters; and an optimization unit that optimizes the decoding parameters to minimize a classification error between the label feature quantity and the label information by using the label feature quantity.

Abstract: A noise determination apparatus includes an acquisition unit and a determination unit. The acquisition unit acquires n-branch tree structure data. The n-branch tree structure data includes information on the presence or absence of a point for a divided region obtained by dividing, by an n-branch tree structure including a plurality of layers, a spatial region represented by point cloud data including information on a position of a point, and includes, for the divided region including only one point, a sign representing the point instead of information on a divided region in a layer lower than the divided region. The determination unit determines a point represented by the sign to be noise in a layer higher than a predetermined layer.

Abstract: A data update method for replacing at least part of encoded point cloud data in which presence or absence of a point is represented by a multi-tree structure for an individual divided region of a region indicated by point cloud data without decoding the multi-tree structure includes acquiring replacement destination data, acquiring replacement source data, which is the encoded point cloud data of the individual divided region corresponding to data to be replaced, and replacing the replacement source data with the replacement destination data.

Abstract: One aspect of the present invention is a data generation method which generates data based on a predetermined estimation model, the data generation method includes a generation step of generating data estimated as a predetermined label by the estimation model and provided with the predetermined label, and generated data has at least either one of a feature close to data to which a label different from the predetermined label is imparted or a feature different from known data to which the predetermined label is imparted.

Abstract: A parameter optimization method includes extracting a feature vector using input data, acquiring a classification result of the feature vector and a class representative vector of every class serving as a classification target, and optimizing a parameter used in the extracting based on a classification error obtained using correct answer data and the classification result and a distance error between the class representative vectors such that areas of features of the classes in a feature space do not overlap each other.

Abstract: Disclosed is a decoding method for decoding encoded data obtained by encoding a cube that includes three-dimensional point cloud data. The decoding method includes a step of obtaining point occupation states of solids obtained after division in one direction, from the encoded data, wherein the direction of the division is determined based on correlation among the point occupation states of the solids after the division. Another decoding method and a decoding apparatus are also disclosed.

Abstract: A decoding method executed by a decoding device for decoding encoded data from point cloud data includes: acquiring an occupancy code represented in an N (1?N?8)-ary tree structure by decoding the encoded data; repeating a process of generating an N-ary tree structure until a block has a predetermined size, the process being a process in which a cube configured to include all the point cloud data is divided into eight blocks and thereafter when a bit of an occupancy code corresponding to a divided block is 1, the block is further divided; and acquiring, as coordinates of a point in the point cloud data, coordinates of a block of the predetermined size where a bit of an occupancy code corresponding to the block of the predetermined size is 1. In the repeating, the decoding device determines a value of N of the N-ary tree structure for a block to be divided on a basis of an inclusion relation between the block to be divided and a non-occupancy region set in advance.

Abstract: A deep feature generation unit (20) inputs an inference image from an input layer (21) of a neural network, performs forward propagation in the neural network, and outputs a plurality of frame images that each include a channel image and are aligned in a predetermined first sequence as intermediate output values from an intermediate layer (22), which is a predetermined layer that is not an output layer of the neural network. A rearrangement unit (30) rearranges the frame images aligned in the first sequence into frame images in a second sequence based on a predetermined rearrangement sequence from the first sequence to the second sequence, such that a total of degrees of similarity between adjacent frame images in the second sequence is greater than a total of degrees of similarity between adjacent frame images in the first sequence.

Abstract: An encoding device includes a division unit that acquires three-dimensional data representing positions of a plurality of points distributed along a surface of an object in a three-dimensional space and divides a parent space including the points in the three-dimensional space into a plurality of child spaces and an encoding unit that changes, based on a position of a target space, which is the child space, to which a sign representing whether the points are included is allocated, according to whether the points are included in a first child space adjacent to the target space, processing for allocating the sign to the target space and a second child space adjacent to the target space.

Abstract: A coding apparatus is a coding apparatus for coding an original image and includes: a division section that divides the original image into blocks that are a plurality of areas to acquire the plurality of blocks; a determination section that, for each of the blocks, determines whether or not to determine the block as an interpolation target; and a substitution section that substitutes a value of a pixel included in the block determined as the interpolation target, with a value that decreases a code amount of the block determined as the interpolation target. The determination section determines whether or not to determine the area that is a target of the determination as an area that is the interpolation target, using an evaluation based on an accuracy of prediction of an image of the block by intra prediction or inter prediction and a degree of the area that is the interpolation target being a generated one.

Abstract: An encoding device includes: an association processing unit that associates first encoded data which is encoded data of an original image with second encoded data which is encoded data of a decoded image which is a result of decoding of the encoded data of the original image; and an encoding unit that encodes, based on the result of associating the first encoded data with the second encoded data, a target image which is an image to be encoded.

Abstract: A generation apparatus includes an interpolation unit that generates, from a moving image including a plurality of frames, an interpolated frame in which some regions in one or more frames included in the moving image are interpolated, and a discrimination unit that discriminates whether a plurality of input frames is interpolated frames in which some regions in the plurality of input frames is interpolated. The discrimination unit includes a temporal direction discrimination unit that discriminates time-wise the plurality of input frames, a spatial direction discrimination unit that discriminates space-wise the plurality of input frames, and an integrating unit that integrates discrimination results from the temporal direction discrimination unit and the spatial direction discrimination unit.

Abstract: In a code amount estimating method, when encoding quantized values of coefficients of a larger-sized orthogonal transformation than an orthogonal transformation size assigned to a variable length encoding table, the quantized values are rearranged in a one-dimensional form, so as to obtain run-level sets. The number of groups is computed based on a proportion between an orthogonal transformation area corresponding to the orthogonal transformation size assigned to the variable length encoding table and an orthogonal transformation area for an encoding target. The Run-Level sets are classified into groups having the number of groups. Each Run is divided by the number of groups, and the obtained quotient is set as Run. A code length of each Run-Level set in each group is determined by referring to the variable length encoding table. The amount of generated code is estimated to be the total sum of the code lengths of all groups.

Abstract: A quantization control method used in a video encoding which encodes a video image and performs control for making an encoding bit rate approach a predetermined target bit rate. The quantization control method includes measuring a differential amount of code between a target amount of code and an amount of generated code; determining whether or not a predetermined condition has occurred; determining, when it is determined that the condition has occurred, a variation for an amount of feedback which increases or decreases a quantization step size, and changing the amount of feedback based on the determined variation; and increasing or decreasing the quantization step size based on the measured differential amount of code and the changed amount of feedback. If there are a plurality of the predetermined conditions, a final variation for the amount of feedback may be determined by applying a specific operation to variations for the amount of feedback, which are determined for the individual conditions.

Abstract: A video bitrate control method and apparatus that control a generated bitrate of a picture to be encoded, based on an initial value of an allocation bitrate that is previously given to each picture are provided, wherein a difference between an allocation bitrate and an actually generated bitrate of a picture that has been encoded is obtained as an error bitrate, one of a maximum value and a minimum value of the allocation bitrate is selected for a plurality of subsequent pictures, as an allowable correction range, based on the sign of the error bitrate, the maximum value and the minimum value of the allocation bitrate being calculated based on the initial value of the allocation bitrate and a previously given constant, a difference between the allowable correction range and the allocation bitrate is obtained for the plurality of pictures, as an allowable variation bitrate, the ratio of the sum of allowable variation bitrates and the error bitrate is obtained as an update rate, a variation bitrate for the allo

Abstract: A motion vector search method used in video encoding which performs motion-compensated prediction. The method includes inputting a motion vector which has been searched for with respect to an encoding target block; computing a motion vector having a minimum overhead cost which minimizes the amount of code generated for the motion vector of the encoding target block; limiting a search area based on the input motion vector and the motion vector having the minimum overhead cost; and searching for a motion vector by searching only the limited search area. It is possible to compute a predicted vector of the encoding target block based on a motion vector of an already-encoded block in the vicinity of the target block, and determine the computed predicted vector to be the motion vector having the minimum overhead cost.

Abstract: An encoding bit-rate control method used in video encoding in which intraframe prediction and interframe prediction are switchably used. The method includes measuring an amount of code generated for quantized information and an amount of code generated for non-quantized information for the picture which has been encoded; computing a ratio of the amount of code generated for the non-quantized information to the total amount of generated code, based on the above measured amounts of code; and determining a target amount of code of an encoding target picture by using the computed ratio of the amount of code generated for the non-quantized information.