Hierarchical V-MESH Displacement Coding for Lower 3D Mesh Bandwidth
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for compressing 3D point clouds and polygonal meshes require specialized hardware and result in high bandwidth usage due to the large amount of data involved, especially in dynamic scenarios where frames are transmitted at high rates.
Innovation Solution
Implement hierarchical coding at the subdivision level for vertex mesh (V-MESH) by creating additional subdivisions from reconstructed subdivided meshes and encoding each displacement field into a bitstream as it is generated, allowing for improved compression and reconstruction of mesh-frames.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If current compression methods are used for 3D multimedia data, then transmission can be achieved, but bandwidth requirements are significant and specialized hardware is needed
Solution Approach 1:
The patent segments the mesh representation into multiple levels of detail (LOD) using hierarchical subdivision. The mesh is divided into coarse geometry (base mesh) and fine geometry (displacement fields), allowing selective transmission based on required precision. This segmentation reduces overall data volume and bandwidth requirements while enabling reconstruction without specialized hardware.
Solution Approach 2:
The patent transforms the 3D mesh compression problem into a multi-dimensional solution by introducing temporal dimension (motion vectors for dynamic content) and hierarchical dimension (multiple LOD levels). Displacement fields are encoded in screen-space coordinates and transformed into mesh space, adding mathematical transformation dimensions that enable efficient compression and hardware-independent processing.
2Loss of energy
If high compression efficiency is achieved for dynamic 3D content, then bandwidth is reduced, but real-time processing becomes more challenging
Solution Approach 1:
The patent performs preliminary action by pre-computing the hierarchical mesh structure and displacement fields during encoding. Motion vectors and displacement data are prepared in advance and organized into prediction units, allowing the decoder to simply reconstruct rather than compute complex transformations in real-time. This shifts computational burden from decode to encode side, enabling real-time playback.
Solution Approach 2:
The patent introduces motion vectors as an intermediary element that bridges temporal frames. Instead of transmitting complete mesh data for each frame, motion vectors serve as mediators that predict vertex displacements between frames. The displacement fields act as intermediaries between screen-space coordinates and mesh-space reconstruction, enabling efficient real-time processing through incremental updates.
Data Source
AI summary
An apparatus includes a communication interface and a processor operably coupled to the communication interface. The processor is configured to subdivide an intermediate mesh-frame, created from a reconstructed first displacement field and a first subdivided mesh-frame, to create a second subdivided mesh-frame. The processor is also configured to generate a second displacement field based on the second subdivided mesh-frame and an original mesh-frame. The processor is further configured to encode the second displacement field into a bitstream.