Stereo Depth Mapping with SNR-Based Fallback for Parallax Correction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing MR systems face challenges in generating accurate depth maps due to parallax errors and poor signal-to-noise ratios (SNR) in stereo camera images, particularly in low light or low ambient conditions, leading to low-quality depth maps that affect the user's perception of the real-world environment.

Innovation Solution

The system computes a smoothness penalty against the smoothness term of a cost function based on the SNR of texture images to improve depth map generation, adjusting the stereo matching algorithm to generate usable depth maps even in low-quality image conditions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If full parallax correction using stereo depth matching is applied, then depth map accuracy is improved, but system reliability deteriorates in low SNR conditions

Engineering Contradiction:
Improvedepth map accuracyVSAvoidsystem reliability in low SNR conditions
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The system dynamically adjusts the depth mapping approach based on real-time SNR conditions. When SNR is high, full stereo depth matching is used for maximum accuracy. When SNR is low, the system transitions to planar reprojection to maintain reliability, thus making the system adaptive to changing environmental conditions rather than static

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system changes the operational parameters of the depth mapping algorithm based on SNR measurements. By monitoring signal quality and adjusting the processing approach (from full parallax correction to planar reprojection), the system optimizes performance across varying environmental conditions while maintaining both accuracy and reliability

Inventive Principle:
Principle #35Parameter changes

2Loss of information

If stereo depth matching is performed on low quality images, then depth information is obtained, but depth map quality deteriorates due to poor SNR

Engineering Contradiction:
Improvedepth information availabilityVSAvoiddepth map quality
Core Design Contradiction:
Loss of informationVSManufacturing precision

Solution Approach 1:

The system introduces SNR measurement as an intermediary assessment step between image capture and depth mapping. This intermediary evaluation allows the system to select the appropriate processing pathway (full stereo matching or planar reprojection) based on image quality, thus preventing degradation of depth map quality while preserving depth information availability

Inventive Principle:
Principle #24Intermediary (Mediator)

3Adaptability or versatility

If gradual fallback from full parallax correction to planar reprojection is implemented, then system adaptability is improved, but device complexity increases

Engineering Contradiction:
Improvesystem adaptability to SNR conditionsVSAvoiddevice complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The depth mapping system is segmented into distinct processing pathways: full stereo depth matching for high SNR conditions and planar reprojection for low SNR conditions. This segmentation allows the system to adapt to different environmental conditions while keeping each individual pathway relatively simple, managing overall complexity through modular design

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP4150579B1Gradual fallback from full parallax correction to planar reprojection
Publication Date: 2025.11.12 MICROSOFT TECHNOLOGY LICENSING LLC
  • EP4150579B1 patent drawingFigure 1
  • EP4150579B1 patent drawingFigure 2
  • EP4150579B1 patent drawingFigure 3

AI summary

Improved techniques for generating depth maps are disclosed. A stereo pair of images of an environment is accessed. This stereo pair of images includes first and second texture images. A signal to noise ratio (SNR) is identified within one or both of those images. Based on the SNR, which may be based on the texture image quality or the quality of the stereo match, there is a process of selectively computing and imposing a smoothness penalty against a smoothness term of a cost function used by a stereo depth matching algorithm. A depth map is generated by using the stereo depth matching algorithm to perform stereo depth matching on the stereo pair of images. The stereo depth matching algorithm performs the stereo depth matching using the smoothness penalty.