Tessellated Mesh Reprojection Using Precomputed Depth Patterns
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing late-stage reprojection methods in augmented/mixed reality scenarios suffer from high computational costs and visual artifacts due to the use of adaptively tessellated grids, which are inefficient and power-intensive, especially in remote rendering scenarios with limited computational resources.
Innovation Solution
The method employs an indexed tessellated mesh computed using a database of pre-computed patterns, allowing for efficient tessellation and rendering by storing the mesh patterns offline, reducing computational overhead and latency, and using integer subdivision factors to minimize oversampling and artifacts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If adaptively tessellated grids are used for late-stage reprojection, then manufacturing precision is improved, but use of energy increases and device complexity worsens
Solution Approach 1:
The patent pre-computes and stores tessellation patterns in a database before runtime. During late-stage reprojection, the system retrieves pre-computed patterns based on depth complexity maps rather than computing tessellation in real-time, significantly reducing energy consumption while maintaining reprojection accuracy
Solution Approach 2:
The system uses depth complexity maps to identify regions requiring different levels of tessellation detail. High-complexity regions receive finer tessellation patterns from the database, while low-complexity regions use coarser patterns, optimizing energy usage by applying computational resources only where necessary
2Manufacturing precision
If adaptively tessellated grids are used for late-stage reprojection, then manufacturing precision is improved, but device complexity worsens
Solution Approach 1:
Tessellation patterns are pre-computed and stored in a database structure with indexing mechanisms. The runtime system only needs to query and retrieve appropriate patterns based on depth complexity maps, avoiding complex real-time tessellation algorithms and reducing device complexity
Solution Approach 2:
The system uses indexed copies of pre-computed tessellation patterns stored in a database. Instead of generating unique tessellation structures for each reprojection operation, the system retrieves and applies appropriate pre-stored patterns, simplifying the computational system
3Adaptability or versatility
If real-time tessellation computation is performed, then adaptability is improved, but productivity worsens
Solution Approach 1:
Tessellation patterns for various depth complexity levels are pre-computed and stored in an indexed database. During reprojection, the system quickly retrieves appropriate patterns using depth complexity maps as indices, maintaining adaptability to different scene complexities while achieving high reprojection speeds
Solution Approach 2:
The system adapts tessellation detail to local depth complexity requirements by querying the database with depth complexity map data. Each region receives appropriately detailed pre-computed patterns, maintaining adaptability without the computational overhead of real-time tessellation generation
4Manufacturing precision
If high-resolution tessellation is applied uniformly, then manufacturing precision is improved, but loss of substance increases
Solution Approach 1:
The system uses depth complexity maps to identify regions requiring high versus low tessellation detail. Pre-computed patterns of varying resolution are retrieved from the database and applied selectively to different regions, ensuring high image quality where needed while conserving computational resources in less critical areas
Data Source
AI summary
A method of late stage reprojection comprises accessing a depth buffer and a color image, accessing a depth complexity map, computing an indexed tessellated mesh from the depth complexity map using a database of pre-computed patterns and reprojecting the indexed tessellated mesh to correct the color image.