Geometry Optimization for Real-Time 3D Rendering

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing 3D computer graphics methods require significant computational resources, especially for real-time rendering of composite products, due to the complexity of models and inefficiencies in occlusion culling algorithms, leading to decreased frame rates and user experience.

Innovation Solution

A method that subdivides computer models into parts based on spatial relationships between components, allowing for view-independent optimization by determining which parts are hidden and can be removed without using complex ray tracing, thereby reducing computational load and only re-running optimization when structural changes occur near the affected area.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If occlusion culling algorithms are used to reduce computational load, then rendering efficiency is improved, but the algorithms are slow and need to be re-run every time the view changes, increasing computational load

Engineering Contradiction:
Improverendering efficiencyVSAvoidcomputational time for re-running optimization
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

The model is divided into multiple components with defined spatial relationships. Instead of performing occlusion culling on the entire model, the system segments the rendering optimization into component-level operations, allowing selective re-optimization of only those components affected by view changes or structural modifications.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Spatial relationships between components are pre-determined and stored before rendering. This preliminary calculation of component positions and orientations allows the system to quickly identify which components need re-optimization without performing full occlusion culling when views change, reducing repeated computational overhead.

Inventive Principle:
Principle #10Preliminary action

2Measurement precision

If complex occlusion culling algorithms are used to accurately determine hidden surfaces, then rendering accuracy is improved, but computational resources increase significantly

Engineering Contradiction:
Improveocclusion detection accuracyVSAvoidcomputational resources
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent extracts and utilizes the spatial relationship information between components as a separate, reusable data structure. By extracting this information once and storing it, the system avoids repeatedly performing complex occlusion detection calculations, reducing computational resource consumption while maintaining accurate hidden surface determination.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

Instead of re-calculating occlusion relationships from scratch for each view change, the system creates and stores copies of spatial relationship data between components. These pre-computed spatial relationships are then referenced during rendering to quickly determine which parts to hide, avoiding repeated expensive calculations.

Inventive Principle:
Principle #26Copying

3Measurement precision

If the entire model is re-optimized every time the view changes, then rendering accuracy is maintained, but frame rate decreases

Engineering Contradiction:
Improverendering accuracyVSAvoidframe rate
Core Design Contradiction:
Measurement precisionVSSpeed

Solution Approach 1:

The system dynamically determines which components require re-optimization based on view changes and structural modifications. Instead of statically re-optimizing the entire model, the system adapts its optimization scope to only include affected components, maintaining rendering accuracy while improving frame rate through selective processing.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

Different optimization strategies are applied to different components based on their spatial relationships and visibility. Components that are occluded or structurally related to changed areas receive detailed optimization, while visible or independent components use simpler rendering paths, ensuring local quality where needed without sacrificing overall performance.

Inventive Principle:
Principle #3Local quality

Data Source

PatentUS9230360B2Connectivity depended geometry optimization for real-time rendering
Publication Date: 2016.01.05 LEGO AS

AI summary

Disclosed is a computer-implemented method for visualizing/rendering a computer model, the computer model comprising a plurality of components wherein the method comprises: storing a plurality of parts of at least one of the plurality of components, each part including geometry information for visualizing/rendering the part, storing information indicative of the position and orientation of a plurality of components in a coordinate system, determining a spatial relationship between individual components, determining for the at least one component a number of stored parts to use for visualizing/rendering the component, based on the determined spatial relationship between the individual components. By using the spatial relationship between the components to determine the number of parts to use for visualizing/rendering a component, knowledge about the geometry of individual components and the connections of the components with other components can be used to find hidden parts that can be removed.