Point cloud data frames compression

A method for decoding a frame of a sequence of frames of point cloud data for rendering a dynamic three-dimensional (3D) scene, wherein each frame comprises a plurality of points, and each point comprises a 3D position and one or more attributes, the method comprising:

• identifying a current point included within a current frame;
• identifying a preceding point included within a preceding frame, wherein the preceding point corresponds to the current point;
• obtaining ordering information defining an order for a set of points included within the current frame, wherein the ordering information is configured to omit information relating to points that were previously encoded using inter-frame predictive encoding, and wherein the omitted information was omitted based on a determination that inter-frame predictively encoded points already have an order as defined in the preceding frame;
• and
• decoding an attribute of the current point by reference to the preceding point using inter-frame prediction decoding to generate decoded data for the current point.

A computer system comprising:

• one or more processors;
• and
• one or more hardware storage devices that store instructions that are executable by the one or more processors to cause the computer system to:
• access a sequence of frames that comprise point cloud data, wherein the sequence of frames includes a first frame and a second frame, wherein the first frame precedes the second frame within an ordering of the sequence of frames;
• identify a first point included within the first frame;
• identify a plurality of points included within the second frame, wherein the plurality of points includes a second point;
• determine that the second point corresponds with the first point in that both the first point and the second point represent common data;
• generating ordering information defining an order for the plurality of points included within the second frame, wherein the ordering information is configured to omit information relating to points that will be subsequently encoded using inter-frame predictive encoding, and wherein omitting the information is performed based on a determination that inter-frame predictively encoded points already have an order as defined in the first frame;
• and
• use inter-frame predictive encoding to encode an attribute of the second point, wherein the inter-frame predictive encoding is performed for the second point by referencing the first point, and wherein encoding the attribute results in generation of encoded data for the second point.