3D Mesh Compression via Patch Projection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing mesh compression methods, such as those extending VPCC, face challenges in maintaining encoding efficiency while preserving subjective image quality, as reducing vertex numbers in meshes can lead to reduced geometry or texture resolution.
Innovation Solution
The proposed solution involves generating a base mesh with fewer vertices than a target mesh, dividing the target mesh into patches, and projecting these patches onto the base mesh. This approach allows for the encoding of meta information including vertex connectivity, while maintaining higher resolution geometry and texture images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If the number of vertices in the mesh is reduced to decrease the code amount for vertex connectivity information, then encoding efficiency is improved, but the resolution of geometry or texture in the reconstructed mesh is reduced, worsening subjective image quality
Solution Approach 1:
The target mesh is divided into multiple patches, each representing a portion of the base mesh. This segmentation allows the geometry and texture information to be encoded separately from the vertex connectivity information of the base mesh, resolving the contradiction by enabling reduced vertex count in the base mesh while preserving high-resolution geometry and texture through the patches.
Solution Approach 2:
The base mesh serves as an intermediary structure with fewer vertices that provides the vertex connectivity framework. The high-resolution geometry and texture information are then mapped onto this base mesh through patches, allowing the base mesh to have reduced vertex count while the patches preserve the high-resolution details needed for quality reconstruction.
2Adaptability or versatility
If vertex connectivity information is encoded separately from geometry images and texture images, then the mesh structure can be represented, but encoding efficiency decreases due to increased code amount
Solution Approach 1:
The invention merges the encoding of vertex connectivity information with the encoding of geometry and texture information by using the base mesh structure. The patches are arranged on a frame image along with the geometry image, allowing all information to be encoded together in a unified structure, thereby improving encoding efficiency while maintaining mesh structure representation.
3Loss of information
If a base mesh with fewer vertices is used to represent the three-dimensional structure, then the code amount is reduced, but the accuracy of shape and texture reconstruction may be compromised
Solution Approach 1:
By segmenting the target mesh into multiple patches that are projected onto the base mesh, the invention preserves high-resolution geometry and texture information in the patches even though the base mesh has fewer vertices. This segmentation allows the base mesh to provide the structural framework while the patches maintain the detailed accuracy needed for faithful reconstruction.
Solution Approach 2:
The patches serve as copies of the high-resolution geometry and texture information from the target mesh. These patches are projected onto the base mesh structure, allowing the base mesh with fewer vertices to be augmented with high-resolution details through the patch copies, thereby maintaining accuracy while reducing overall code amount.
Data Source
AI summary
There is provided an information processing device and a method that allow for suppression of reduction in encoding efficiency while suppressing degradation in subjective image quality. A base mesh with fewer vertices than a target mesh is generated, a plurality of patches are generated by dividing the target mesh and projecting the divided parts on the base mesh, a geometry image is generated by arranging the patches on a frame image, and vertex connectivity information about the base mesh and the geometry image are encoded. The vertex connectivity information and the encoded data of the geometry image are decoded, the number of vertices of the base mesh is increased using the vertex connectivity information, the patches are reconstructed, and reconstructed vertex information is generated using the reconstructed patches. The present disclosure can be applied for example to an information processing device, an electronic apparatus, an information processing method, or a program.