Planar Surface Detection via Sub-Volume Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for processing planar surface information in large-scale physical environments are computationally intensive, leading to battery drain and delays in augmented reality experiences, especially for small mixed reality devices.

Innovation Solution

The environment is partitioned into virtual sub-volumes, with surface representations generated in each sub-volume, allowing for individual processing of planar regions and reducing computational load by only processing relevant sub-volumes.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If existing methods process planar surface information in large-scale environments, then complete surface detection is achieved, but computational load and power consumption increase significantly

Engineering Contradiction:
Improveplanar surface detection accuracyVSAvoidpower consumption
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent divides the large-scale physical environment into multiple smaller sub-volumes or regions. Each sub-volume is processed independently for planar surface detection, reducing the computational burden on any single processing unit. This segmentation allows mobile devices to detect planar surfaces accurately within each region without being overwhelmed by the entire environment's data processing requirements, thereby reducing power consumption while maintaining detection accuracy.

Inventive Principle:
Principle #1Segmentation

2Measurement precision

If existing methods process all environment data for planar surface detection, then comprehensive surface information is obtained, but processing time and delays increase

Engineering Contradiction:
Improvesurface information completenessVSAvoidprocessing delay
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

By segmenting the environment into sub-volumes, the patent enables parallel processing of multiple regions simultaneously. This reduces the sequential processing time required to analyze the entire environment, as different sub-volumes can be processed concurrently. The system maintains comprehensive surface information by ensuring all sub-volumes are processed, but the overall processing time is reduced through this divide-and-conquer approach.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements a approach where initially only relevant or visible sub-volumes are processed for planar surface detection, rather than processing the entire environment data set. This partial action reduces processing time and delays, while still providing sufficient surface information for augmented reality experiences. Additional sub-volumes can be processed as needed based on user interaction or device capabilities.

Inventive Principle:
Principle #16Partial or excessive action

3Measurement precision

If mobile devices process large-scale environment data, then accurate planar surface detection is achieved, but device performance and battery life deteriorate

Engineering Contradiction:
Improveplanar surface detection accuracyVSAvoiddevice performance
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent segments the processing task into manageable sub-volume units that can be handled by mobile device processors without causing performance degradation. Each sub-volume contains a limited amount of environment data that can be processed efficiently, maintaining detection accuracy while preventing the device from being overwhelmed by the computational demands of processing the entire large-scale environment at once.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP3959589B1Planar surface detection
Publication Date: 2025.06.18 MICROSOFT TECHNOLOGY LICENSING LLC
  • EP3959589B1 patent drawingFigure 1A
  • EP3959589B1 patent drawingFigure 1B
  • EP3959589B1 patent drawingFigure 1C

AI summary

The described implementations relate to processing of an environment using a plurality of sub-volumes, and specifically to generating surface representations in the plurality of sub-volumes for individual processing. One example can identify planar fragments within the plurality of sub-volumes. The example can determine that various planar fragments constitute part of a contiguous planar surface and should be aggregated. The example can also output data representing the contiguous planar surface formed from the aggregated planar fragments.