Replicated Dot Maps for Depth Detection in Mixed Reality

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional mixed-reality systems face challenges in accurately determining depth, particularly for smooth or textureless surfaces, as existing depth detection methods fail to distinguish between different parts of these surfaces, leading to inaccurate depth maps and a compromised user experience.

Innovation Solution

The use of a structured projection pattern, specifically an infrared (IR) dot pattern, is projected onto the environment, allowing a stereo camera system to perform depth detection by matching pixel patches between captured images, enabling accurate depth calculation even for smooth surfaces through stereo matching and structured light approaches.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional passive stereo depth detection systems are used, then the system structure remains simple, but depth detection accuracy deteriorates for smooth or textureless surfaces

Engineering Contradiction:
Improvedepth detection accuracyVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent introduces an infrared structured light projector as an intermediary device that projects a dot pattern onto the environment. This mediator provides artificial texture information on smooth surfaces, enabling the stereo camera system to detect depth accurately without requiring complex hardware changes to the existing mixed-reality system.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The system performs preliminary action by projecting the infrared dot pattern onto the environment before depth detection occurs. This pre-applied structured light pattern creates distinguishable features on smooth surfaces in advance, allowing the stereo cameras to capture and process depth information effectively during the subsequent detection phase.

Inventive Principle:
Principle #10Preliminary action

2Measurement precision

If conventional dot patterns are used for depth detection, then depth information can be obtained, but computational burden increases due to need to distinguish unique patterns across large areas

Engineering Contradiction:
Improvedepth detection capabilityVSAvoidcomputational efficiency
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent divides the dot pattern into multiple replicated sub-patterns, each containing unique arrangement characteristics. This segmentation allows the system to process smaller, manageable pattern units rather than treating the entire large-area pattern as a single complex structure, thereby reducing computational burden while maintaining depth detection accuracy.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system uses replicated copies of sub-patterns distributed across the environment. Each sub-pattern is a copy with consistent internal structure but unique positioning and arrangement. This copying approach enables the system to recognize patterns through replication rather than processing entirely unique complex patterns, significantly improving computational efficiency.

Inventive Principle:
Principle #26Copying

3Area of stationary object

If unique dot patterns are projected across large areas, then depth detection coverage is improved, but pattern recognition difficulty increases

Engineering Contradiction:
Improvedepth detection coverageVSAvoidpattern recognition difficulty
Core Design Contradiction:
Area of stationary objectVSDifficulty of detecting and measuring

Solution Approach 1:

The patent segments the large-area dot pattern into multiple smaller sub-patterns that are replicated across the coverage area. Each sub-pattern maintains a consistent internal structure that is easier to recognize, while the collection of replicated sub-patterns collectively provides extensive depth detection coverage. This segmentation reduces pattern recognition difficulty compared to using a single large unique pattern.

Inventive Principle:
Principle #1Segmentation

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

This method significantly improves depth detection accuracy for mixed-reality environments, reducing computational burden and resource allocation, while enhancing user experience by providing precise placement of virtual objects in relation to real-world objects.

Implementation Method 1

an infrared (IR) illuminator, wherein the dot pattern includes a replicated sub-pattern

Methodology Applied
Scientific EffectInfrared radiation: Infrared Radiation

Implementation Method 2

The structured projection pattern is reflected off of the object(s), and the camera system obtains at least one image of the reflected structured projection pattern

Methodology Applied
Scientific EffectLight reflection: Reflection

Data Source

PatentEP3769035B1Replicated dot maps for simplified depth computation using machine learning
Publication Date: 2023.11.15 MICROSOFT TECHNOLOGY LICENSING LLC
  • EP3769035B1 patent drawingFigure 1
  • EP3769035B1 patent drawingFigure 2
  • EP3769035B1 patent drawingFigure 3

AI summary

Disclosed embodiments include methods and systems for utilizing a structured projection pattern to perform depth detection. In some instances, the structured projection pattern forms a dot pattern, which is projected by an infrared (IR) illuminator, wherein the dot pattern includes a replicated sub-pattern having a predefined height and width. The sub-pattern is replicated in at least one direction such that the dot pattern comprises a plurality of replicated sub-patterns that are adjacently positioned.