Methods and devices for coding point cloud data using volume trees and predicted-point trees. In one embodiment of the disclosure, a method of encoding a point cloud data to generate a bitstream of compressed point cloud data representing a three-dimensional location of a physical object is provided, the point cloud data being located within a volumetric space. The method includes compressing a first part of the point cloud data represented by a first tree of a first type; determining for a given node of the first tree if an assignation to a second type of tree is enabled, said given node still being processed for the first tree; when the assignation is enabled, compressing a second part of the point cloud data represented by a second tree of the second type wherein, features associated with a root node of the second tree are at least partially obtained from the given node.

A control circuit facilitates compressing source field information (such as, for example, information that comprises a scalar field and or a vector field) having a corresponding initial space for a given object into a corresponding compact representation. That source field information may comprise a three-dimensional object represented as a two-dimensional manifold embedded in Euclidean three-dimensional space approximated via polygon mesh. This can comprise subdividing the initial space into a plurality of subspaces and generating a fixed-dimensionality vector representation for each field that corresponds to one of the subspaces. These teachings can then provide for inputting the fixed-dimensionality vector representations and query point coordinates corresponding to each of the subspaces to a field estimator neural network (such as, but not limited to, a neural network configured as an encoder-decoder machine learning model) trained to output corresponding field values.

A system and method of determining synthetic dual-pixel data, performing deblurring, predicting dual pixel views, and view synthesis. The method including: receiving an input image; determining synthetic dual-pixel data using a trained artificial neural network with the input image as input to the trained artificial neural network, the trained artificial neural network includes a latent space encoder, a left dual-pixel view decoder, and a right dual-pixel view decoder; and outputting the synthetic dual-pixel data. In some cases, determination of the synthetic dual-pixel data can include performing reflection removal, defocus deblurring, or view synthesis.

A three-dimensional data encoding method includes: generating predicted position information using position information on three-dimensional points included in three-dimensional reference data associated with a time different from a time associated with current three-dimensional data; and encoding position information on three-dimensional points included in the current three-dimensional data, using the predicted position information.

An example device for coding point cloud data includes a memory configured to store point cloud data; and one or more processors implemented in circuitry and configured to: determine at least one of 1) that a node of an octree of the point cloud data is not inter predictable or 2) that angular mode is enabled for the node; in response to determining the at least one of 1) that the node is not inter predictable or 2) that angular mode is enabled for the node, determine an inferred direct coding mode (IDCM) mode for the node; and code occupancy data of the node using the determined IDCM mode.

An example method of encoding a point cloud includes, responsive to determining that a first point of the point cloud is a first point in a first group of points of one or more groups of points of the point cloud, encoding, in a bitstream, one or more syntax elements related to inter prediction for the first group of points. The example method may further include, responsive to determining that a second point of the point cloud is included in the first group of points but is not the first point in the first group of points, skip re-encoding, for the second point, the one or more syntax elements related to inter prediction for the first group of points.

The present disclosure relates to an information processing device and a method capable of suppressing a reduction in encoding efficiency of point cloud data. As for a point cloud representing an object having a three-dimensional shape as a point group, position information of a point to be processed is predicted on the basis of position information of a reference point, position information of a prediction point is generated, a difference between the generated position information of the prediction point and the position information of the point to be processed is derived, the derived difference is encoded, and a bitstream is generated. The present disclosure may be applied to, for example, an information processing device, an electronic device, an information processing method, a program or the like.

Provided are a method and apparatus for decoding video signal based on weighted prediction. The method may include determining an inter prediction mode of a current block, deriving motion information of a current block according to the inter prediction mode, obtaining a first prediction block of the current block based on the motion information, and obtaining a second prediction block of the current block by applying at least one of a weight, an offset, or a first variable for explicit weighted prediction to the first prediction block.

In the field of computer vision, a decoding method, an encoding method, a decoder, and an encoder based on point cloud attribute prediction are provided. The decoding method includes: parsing a code stream of a point cloud to obtain reconstructed information of position information of a target point; selecting candidate points of the target point from decoded points in the point cloud; selecting neighbor points from the candidate points based on the reconstructed information of the position information of the target point; determining a predicted value of attribute information of the target point by using attribute values of the neighbor points; and obtaining a decoded point cloud based on the predicted value of the attribute information of the target point. Neighbor points with attributes similar to that of a target point are selected where possible to predict attribute information of the target point, thereby reducing the prediction complexity.

A point cloud data reception method according to embodiments can comprise the steps of: receiving a bitstream including point cloud data; and decoding the point cloud data. A point cloud data transmission method according to embodiments can comprise the steps of: encoding point cloud data; and transmitting a bitstream including the point cloud data.