Autonomous Gating for Noise Reduction in Direct Time-of-Flight Depth Sensing

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional depth sensing technologies, particularly direct time-of-flight configurations, face challenges in accurately mapping environments due to background noise, which affects the accuracy of depth estimation in virtual and augmented reality systems.

Innovation Solution

The depth camera assembly employs time-of-flight depth sensing with single-photon avalanche diodes and uses time gating to reduce noise, along with a controller that dynamically allocates light pulses based on confidence measures from recent frames to improve depth estimation accuracy, optimizing the number of pulses across the detector to minimize noise and power consumption.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If direct time-of-flight depth sensing is used to achieve high depth estimation accuracy, then measurement precision is improved, but background noise increases depth estimation errors

Engineering Contradiction:
Improvedepth estimation accuracyVSAvoidbackground noise
Core Design Contradiction:
Measurement precisionVSObject-affected harmful factors

Solution Approach 1:

The patent applies periodic pulsed illumination instead of continuous illumination, sending light in discrete time-separated pulses. This periodic action allows the detector to sample only during specific time windows when reflected photons are expected, thereby separating signal from background noise and improving depth estimation accuracy while maintaining low power consumption.

Inventive Principle:
Principle #19Periodic action

Solution Approach 2:

The patent performs preliminary actions by pre-configuring time-gated detectors with specific integration time windows before each pulse arrival. The system also performs preliminary background noise characterization and uses confidence measures from previous frames to adaptively adjust exposure parameters, preparing the system in advance to optimize signal detection while minimizing noise impact.

Inventive Principle:
Principle #10Preliminary action

2Reliability

If the detector is kept active for longer exposure time to capture more photons, then signal detection is improved, but background noise accumulation increases

Engineering Contradiction:
Improvesignal detection capabilityVSAvoidbackground noise
Core Design Contradiction:
ReliabilityVSObject-affected harmful factors

Solution Approach 1:

The detector operates in periodic cycles, being activated only during specific time windows corresponding to expected photon arrival times from reflected pulses. This periodic activation allows the detector to accumulate signal photons efficiently while remaining inactive during periods when only background noise would be detected, thereby improving signal detection capability without proportionally increasing noise accumulation.

Inventive Principle:
Principle #19Periodic action

Solution Approach 2:

The patent implements dynamic adjustment of detector integration time windows and exposure parameters based on real-time conditions. The system adapts the timing and duration of detector activation dynamically, extending integration windows when signal strength is weak but maintaining short windows when background noise would dominate, thereby optimizing the balance between signal detection and noise rejection.

Inventive Principle:
Principle #15Dynamics

3Measurement precision

If more light pulses are allocated to improve depth estimation accuracy, then measurement precision is improved, but power consumption increases

Engineering Contradiction:
Improvedepth estimation accuracyVSAvoidpower consumption
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent applies local quality by differentiating pulse allocation across different spatial regions and detector units. Instead of uniformly increasing pulses everywhere, the system allocates additional pulses selectively to regions where depth estimation confidence is low or where measurement precision needs improvement, while maintaining reduced pulse rates in regions with already high confidence, thereby improving overall measurement precision without proportionally increasing total power consumption.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The system dynamically changes operational parameters including pulse frequency, integration time, and detector gain based on confidence measures from previous frames and real-time signal quality assessment. By adapting these parameters rather than simply increasing pulse count, the system optimizes depth estimation accuracy while minimizing power consumption, adjusting the balance between measurement precision and energy use based on environmental conditions.

Inventive Principle:
Principle #35Parameter changes

4Measurement precision

If time gating is used to reduce background noise, then measurement precision is improved, but device complexity increases

Engineering Contradiction:
Improvedepth estimation accuracyVSAvoiddetector control complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent implements segmentation by dividing the detector into multiple independently controllable units or regions, each with its own time-gating capability. This segmentation allows different time windows to be applied to different spatial regions, enabling precise noise rejection for each region while maintaining overall system manageability. The segmented architecture simplifies the control logic compared to attempting global time gating, as each segment can be independently optimized.

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 approach enhances the accuracy of depth estimation by reducing noise and power consumption, allowing for more precise mapping of environments in virtual and augmented reality systems, improving the interaction of virtual objects with real-world surfaces.

Implementation Method 1

single-photon detectors are used, such as single-photon avalanche diodes (SPADs)

Methodology Applied
Scientific EffectPhotoelectric Effect: Photoelectric Effect

Implementation Method 2

Direct time-of-flight (dTOF) depth sensing configurations measure a roundtrip travel time of photons

Methodology Applied
Scientific EffectTime of Flight: Time of Flight

Data Source

PatentUS11637978B1Autonomous gating selection to reduce noise in direct time-of-flight depth sensing
Publication Date: 2023.04.25 META PLATFORMS TECHNOLOGIES LLC
  • US11637978B1 patent drawing
  • US11637978B1 patent drawing
  • US11637978B1 patent drawing

AI summary

A depth camera assembly (DCA) includes a direct time of flight system for determining depth information for a local area. The DCA includes an illumination source, a camera, and a controller. In some embodiments, the controller uses previous image frames to determine confidence measurements, and selectively adjusts a number of pulses from the illuminator in a subsequent frame based on the determined confidence values. In some embodiments, the sensor uses autonomous gating, and the depth system includes a depth recovery pipeline which provide depth map estimates from the autonomous gated measurements.