VR AR 3D Object Mask Segmentation for Non-Intersecting Meshes

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current virtual reality and augmented reality systems face performance issues in rendering 3D objects, particularly with overlapping topologies, resulting in inadequate resolution and inaccurate mesh creation, leading to intersecting meshes that fail to distinguish between separate 3D objects.

Innovation Solution

A method involving the creation of masks based on 3D object types, assignment of voxels and visual markers, and geodesic distance calculations to generate accurate 3D meshes for VR/AR rendering, allowing for clear visualization of complex anatomies by determining an outer mask and modifying visual markers for each vertex based on the shortest geodesic distance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If traditional mesh generation methods are used for 3D objects with overlapping topologies, then the rendering process is simple, but the meshes intersect and fail to distinguish between separate objects

Engineering Contradiction:
Improvemesh accuracyVSAvoidmesh generation complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent segments the 3D object into multiple masks, where each mask corresponds to a distinct 3D object or region. This segmentation allows the system to process and render each object separately, preventing mesh intersections and maintaining topological accuracy for overlapping structures.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces masks as intermediary structures between the 3D objects and the final mesh representation. These masks serve as mediators that preserve the spatial relationships and topological information of overlapping objects, enabling accurate mesh generation without direct intersection of meshes.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If high resolution rendering is achieved through substantial sampling and calculations, then the visual quality improves, but the computing resources required increase significantly

Engineering Contradiction:
Improvepixel resolutionVSAvoidcomputing resources
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent performs preliminary actions by pre-processing the 3D objects into masks and pre-calculating geodesic distances between vertices and masks. This preliminary processing reduces the computational burden during actual rendering, as the system can leverage pre-computed data to achieve high resolution without requiring substantial real-time sampling and calculations.

Inventive Principle:
Principle #10Preliminary action

3Productivity

If traditional mesh generation is used for overlapping 3D objects, then the process is fast, but the visualization accuracy is poor and objects are indistinguishable

Engineering Contradiction:
Improverendering speedVSAvoidobject distinction accuracy
Core Design Contradiction:
ProductivityVSManufacturing precision

Solution Approach 1:

The patent applies local quality by assigning different visual markers to different masks based on their spatial relationships and topological properties. This allows the system to maintain fast rendering while achieving high visualization accuracy, as each local region (mask) is rendered with appropriate distinguishing characteristics without requiring complete reprocessing of the entire scene.

Inventive Principle:
Principle #3Local quality

Data Source

PatentEP3545500B1System and method for rendering complex data in a virtual reality or augmented reality environment
Publication Date: 2021.11.03 3D SYSTEMS INC
  • EP3545500B1 patent drawingFigure 1
  • EP3545500B1 patent drawingFigure 2B
  • EP3545500B1 patent drawing

AI summary

Methods for virtual reality and augmented reality rendering of a 3D object are provided. The methods can include segmenting the 3D object such that portions of the 3D object can be identifiable and distinguishable. The methods can include creating masks, identifying relationships between the masks and marking the masks accordingly.