Dynamic Mesh Compression Using PCC and Connectivity Reconstruction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies face challenges in efficiently compressing dynamic meshes, which require significant data resources due to time-varying connectivity and attribute information, impacting storage and transmission efficiency in applications like real-time communications and immersive experiences.
Innovation Solution
Utilizing point cloud compression (PCC) codecs and video coding techniques to decompose dynamic meshes into patches, generate geometry and texture images, and encode these using video coding standards, along with occupancy maps, to create efficient compression frameworks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If dynamic meshes are compressed using traditional methods, then storage and transmission efficiency improves, but connectivity information and attribute information require significant data resources
Solution Approach 1:
The patent segments the mesh compression task into multiple components: geometry compression using PCC, connectivity information compression using video coding techniques, and attribute information compression. This segmentation allows each component to be optimized independently, reducing overall data resources while managing complexity through modular processing.
Solution Approach 2:
The patent applies video coding standards (H.264/AVC, H.265/HEVC, VVC) to compress connectivity information, leveraging existing universal video coding technology for a non-video application. This multi-functional use of video codecs reduces the need for specialized compression algorithms, thereby reducing data resources while keeping the framework manageable.
2Productivity
If mesh compression is performed without interframe prediction, then coding simplicity is maintained, but compression efficiency deteriorates for dynamic meshes
Solution Approach 1:
The patent introduces interframe prediction mechanisms that dynamically adapt to the temporal characteristics of dynamic meshes. The system selectively applies prediction techniques based on motion detection and temporal redundancy analysis, improving compression efficiency for moving or changing mesh structures while maintaining simplicity for static scenes.
Solution Approach 2:
The patent uses motion vectors and reference frame selection as intermediaries to bridge current and previous mesh frames. These intermediaries enable efficient prediction of connectivity and attribute changes without requiring complete re-encoding of each frame, thus improving compression efficiency while controlling coding complexity through standardized video coding tools.
3Measurement precision
If point cloud compression is used for mesh vertices, then geometry representation accuracy is improved, but connectivity information requires additional data resources
Solution Approach 1:
The patent extracts connectivity information from the mesh structure separately from geometry compression. By using PCC for vertex position encoding and independently compressing connectivity using video coding techniques, the system achieves accurate geometry representation while efficiently managing connectivity data resources through dedicated compression streams.
Solution Approach 2:
The patent transforms the 3D mesh connectivity problem into a 2D image coding problem by representing connectivity maps as images that can be compressed using established video standards. This dimensional transformation allows leveraging highly optimized 2D compression algorithms to reduce the data resources required for storing and transmitting mesh connectivity information.
Data Source
AI summary
Aspects of the disclosure provide methods and apparatuses for mesh coding (e.g., compression and decompression). In some examples, an apparatus for mesh coding includes processing circuitry. The processing circuitry decodes, using a point cloud compression (PCC) decoder and from a bitstream, a point cloud that includes points corresponding to at least one of vertices in a mesh or sampling points of polygons in the mesh. The processing circuitry determines, based on the point cloud, estimated connectivity information of the vertices in the mesh, and generate, a reconstructed mesh frame based on the point cloud and the estimated connectivity information.