Joint Spatial-Temporal Mesh Decimation for UV Chart Consistency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing mesh compression standards fail to efficiently handle dynamic meshes with time varying connectivity and attribute maps, leading to poor texture compression due to lack of temporal correlation in UV charts between consecutive frames.
Innovation Solution
Implement joint spatial-temporal mesh decimation by reusing decimated meshes across consecutive frames within a decimation intra-refresh period, ensuring consistent UV charts and efficient texture compression.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If mesh decimation is performed independently for each frame, then processing complexity per frame is reduced, but temporal correlation in UV charts is lost leading to poor texture compression
Solution Approach 1:
The patent merges the decimation process across multiple consecutive frames by selecting one reference frame for decimation and reusing its decimated mesh for other frames within the same decimation intra-refresh period. This combining approach maintains temporal correlation in UV charts while reducing overall processing complexity.
Solution Approach 2:
The patent performs preliminary decimation on a selected reference frame before encoding other frames. By pre-computing the decimated mesh and UV charts for the reference frame, subsequent frames can leverage this pre-computed information, maintaining temporal correlation without repeating the full decimation process.
2Manufacturing precision
If a new decimated mesh is generated for each frame, then mesh accuracy is maintained, but bitrate increases due to redundant encoding of similar structures
Solution Approach 1:
The patent creates copies of the decimated mesh from a reference frame and reuses them for other frames within the decimation intra-refresh period. Instead of generating new decimated meshes for each frame, the same decimated mesh structure is copied and applied, significantly reducing bitrate while maintaining sufficient accuracy for consecutive frames with small motions.
Solution Approach 2:
The decimated mesh generated from one reference frame serves multiple functions for encoding several consecutive frames. This single decimated mesh structure is universally applied across multiple frames within the decimation intra-refresh period, reducing redundancy and bitrate while maintaining encoding efficiency.
3Manufacturing precision
If UV charts are reparameterized for each frame, then texture mapping accuracy is improved, but inter-prediction efficiency between successive frames deteriorates
Solution Approach 1:
The patent merges UV chart parameterization across multiple frames by using the same UV charts from a reference frame for other frames within the decimation intra-refresh period. This merging approach maintains consistent UV coordinate systems across frames, enabling efficient inter-prediction while preserving sufficient texture mapping accuracy.
Solution Approach 2:
The patent performs preliminary UV chart parameterization on the reference frame and reuses these pre-computed UV charts for other frames. By establishing the UV coordinate system in advance for the reference frame, subsequent frames can leverage this pre-established parameterization, maintaining texture mapping accuracy while improving inter-prediction efficiency.
Data Source
AI summary
A method performed by at least one processor in an encoder includes receiving a plurality of frames, each frame including a polygon mesh; selecting one frame from the plurality of frames; performing decimation on the polygon mesh of the one frame to generate a decimated mesh; and encoding, using the decimated mesh, the polygon mesh in at least one frame from the plurality frames excluding the one frame.