Single-Frequency ToF Depth Computation via Stereoscopic Disambiguation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Time-of-flight (ToF) depth sensing systems face challenges in achieving accurate depth computation with single-frequency measurements due to phase ambiguity, and they also suffer from high power consumption and size constraints in mobile devices, which limits their operational efficiency and accuracy.

Innovation Solution

The implementation of stereoscopic de-aliasing using a single-frequency ToF depth computation method, where a structured-light pattern is projected and a virtual camera is used to perform stereoscopic distance computations, reducing the need for multiple frequency captures and thus lowering power consumption and data collection while maintaining accurate depth tracking.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If multiple frequency captures are used to resolve phase ambiguity, then depth measurement precision is improved, but power consumption increases

Engineering Contradiction:

Improvedepth measurement precisionVSAvoidpower consumption

Core Design Contradiction:

Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent segments the depth measurement process into two distinct components: (1) a coarse depth estimate obtained from stereoscopic vision using image disparity, and (2) a fine depth adjustment obtained from single-frequency ToF phase measurement. The coarse estimate resolves the phase wrap ambiguity, allowing the single-frequency ToF to provide accurate depth without requiring multiple frequency captures for disambiguation.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces an intermediary stereoscopic depth estimation as a mediator between the ambiguous single-frequency phase measurement and the final accurate depth value. The stereoscopic disparity provides an intermediate depth estimate that disambiguates the phase wraps, enabling the ToF system to achieve multi-frequency accuracy with single-frequency power consumption.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Measurement precision

If multiple frequency captures are used to resolve phase ambiguity, then depth measurement precision is improved, but device complexity increases

Engineering Contradiction:

Improvedepth measurement precisionVSAvoiddevice complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent makes the ToF sensor perform multiple functions: it provides both the primary depth measurement (via phase measurement) and works in conjunction with stereoscopic vision to resolve ambiguities. This multi-functionality eliminates the need for separate multi-frequency capture systems, reducing overall device complexity while maintaining measurement precision.

Inventive Principle:

Principle #6Universality (Multi-functionality)

3Reliability

If higher frame rates are used for depth sensing, then measurement reliability is improved, but power consumption increases

Engineering Contradiction:

Improvemeasurement reliabilityVSAvoidpower consumption

Core Design Contradiction:

ReliabilityVSUse of energy by moving object

Solution Approach 1:

The patent employs periodic action by alternating between stereoscopic image capture and ToF depth measurement in a time-multiplexed manner. Rather than continuously operating at high frame rates with full multi-frequency ToF captures, the system periodically switches between the two modalities, achieving high effective frame rates for reliable depth sensing while minimizing average power consumption through the efficiency of single-frequency ToF captures.

Inventive Principle:

Principle #19Periodic action

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 approach enables accurate depth computation with reduced power consumption and smaller form factor devices, minimizing motion blur and allowing for precise depth tracking with lower precision requirements, making it suitable for applications like head-mounted display devices.

Implementation Method 1

A ToF sensor has a light source to emit light onto nearby objects. The ToF camera can capture light reflected by surfaces of the objects. By calculating the phase of emitted light from the light source compared to the phase of reflected light returning back from an object is converted into a depth computation

Methodology Applied

Scientific EffectTime of flight: Time of Flight

Implementation Method 2

A ToF sensor has a light source to emit light onto nearby objects. The ToF camera can capture light reflected by surfaces of the objects

Methodology Applied

Scientific EffectLight reflection: Reflection

Data Source

PatentEP3586165B1Single-frequency time-of-flight depth computation using stereoscopic disambiguation

Publication Date: 2023.11.01 MICROSOFT TECHNOLOGY LICENSING LLC

Figure 1
Figure 2
Figure 3

AI summary

Disclosed is a technique for 3D reconstruction through a combination of time-of-flight (ToF) and stereoscopy. Use of a single modulation frequency in ToF computation provides a set of ambiguous distances. To determine a single accurate distance out of this candidate set, a stereoscopic comparison of an image pair provides a disambiguating distance. The stereoscopic comparison uses a stored virtual image of at least part of the emitted light, and the detected image of light reflected from an object. The stereoscopic distance is used to determine which of the multiple accurate distances is correct based on proximity. The closest of the multiple distances to the disambiguating distance is taken as the actual distance.