Adaptive V-PCC Point Cloud Streaming with MPEG-DASH Signaling

The signaling mechanism is segmented into multiple hierarchical levels: MPD-level descriptors for overall point cloud stream identification, AdaptationSet-level descriptors for component grouping, and Representation-level parameters for specific version characteristics. This segmentation allows complex point cloud streaming information to be organized in manageable, modular units that can be processed independently at different stages of the streaming pipeline.

The patent introduces a new dimensional layer of signaling by adding XML elements and descriptors that operate alongside the existing MPEG-DASH structure. This creates a multi-dimensional signaling space where traditional video/audio streaming parameters coexist with point cloud-specific parameters (such as occupancy map configurations, geometry precision levels, and attribute types), enabling comprehensive point cloud adaptation without disrupting existing streaming infrastructure.

The signaling mechanism implements partial action by providing detailed version information only for point cloud components that have multiple representations. For components with a single version, minimal signaling is used. This selective approach reduces overall signaling data volume while maintaining full client selection flexibility where needed, avoiding the overhead of signaling unnecessary information for all components uniformly.

The signaling structure dynamically adapts to the actual content being streamed. When multiple versions of a point cloud component are available, the system activates detailed version signaling with multiple Representation elements. When only a single version exists, the signaling automatically simplifies. This dynamic behavior ensures that signaling data volume matches the actual adaptation needs, preventing unnecessary data transmission.