Sensor Calibration Reference Map for Depth Texture Fusion

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing sensor fusion systems face challenges in aligning depth and texture sensors positioned far apart, leading to inaccuracies and the need for external calibration targets, especially when cost and practicality constraints limit the deployment of depth sensors across the entire field of view.

Innovation Solution

A calibration system that generates a reference map using depth information from multiple perspectives, allowing for mutual registration of texture and depth sensors without co-location, enabling efficient and accurate fusion of data even when sensors are misaligned by over 20 degrees in yaw, pitch, or roll, and reducing the need for external fiducials.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If depth sensors are deployed across the entire field of view to improve alignment accuracy, then measurement precision improves, but device complexity and cost increase

Engineering Contradiction:
Improvealignment accuracyVSAvoidsensor deployment complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system divides the calibration process into two independent stages: first calibrating each sensor individually against the reference map, then performing mutual registration. This segmentation allows each sensor to be calibrated separately using the same reference framework, improving alignment accuracy without requiring complex inter-sensor calibration mechanisms.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The reference map serves as an intermediary between the depth sensor and texture sensor. Instead of directly calibrating between sensors, both sensors are independently registered to the reference map, which mediates their mutual alignment. This eliminates the need for direct sensor-to-sensor calibration and reduces system complexity.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If depth sensors are positioned close to texture sensors to minimize parallax errors, then measurement precision improves, but adaptability worsens due to space constraints

Engineering Contradiction:
Improveparallax error reductionVSAvoidinstallation flexibility
Core Design Contradiction:
Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The reference map acts as a mediator that enables accurate registration between sensors positioned at different locations. By providing a common reference framework, the system can accommodate various installation configurations without compromising alignment precision, thus improving adaptability while maintaining measurement accuracy.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The system transitions from direct spatial co-location to virtual co-location through the reference map. Sensors can be physically separated in three-dimensional space but are virtually aligned through their independent registration to the reference map, effectively adding a computational dimension to the calibration process.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

3Measurement precision

If external calibration targets are used to align sensors, then measurement precision improves, but ease of operation worsens due to additional setup requirements

Engineering Contradiction:
Improvecalibration accuracyVSAvoidcalibration setup simplicity
Core Design Contradiction:
Measurement precisionVSEase of operation

Solution Approach 1:

The system uses the environment itself as the calibration target through natural feature detection. Instead of requiring external fiducial markers, the sensors calibrate by detecting and registering to features that naturally exist in the scene, making the calibration process self-service and eliminating the need for additional calibration equipment.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The system extracts calibration information from the environment itself rather than relying on external calibration targets. By detecting natural features in the scene and using them for registration, the system removes the requirement for separate calibration objects, simplifying the overall calibration process.

Inventive Principle:
Principle #2Taking out (Extraction)

4Measurement precision

If multiple depth sensors are deployed to cover different perspectives, then measurement precision improves, but device complexity and cost increase

Engineering Contradiction:
Improvedepth coverage accuracyVSAvoidsensor system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The reference map serves multiple functions: it acts as a calibration target, a registration framework, and a common reference for multiple sensors. This universal reference enables multiple depth sensors to be calibrated using the same process, reducing system complexity while maintaining comprehensive depth coverage and precision.

Inventive Principle:
Principle #6Universality (Multi-functionality)

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

Enables real-time fusion of depth and texture information across non-co-located sensors, improving accuracy and reducing costs by allowing a single depth sensor to be used for calibration, eliminating the need for depth sensors at every texture sensor location and minimizing reliance on external calibration targets.

Implementation Method 1

transmitting a first light beam from the first perspective, transmitting a second light beam from the second perspective

Methodology Applied
Scientific EffectLight reflection: Reflection

Implementation Method 2

The depth information can be obtained using a time-of-flight (ToF) measurement using a light detection and ranging (LIDAR) imaging system

Methodology Applied
Scientific EffectTime of flight measurement: Time of Flight

Data Source

PatentEP3845922A1Calibration system for combined depth and texture sensor
Publication Date: 2021.07.07 OUTSIGHT
  • EP3845922A1 patent drawingFigure 1
  • EP3845922A1 patent drawingFigure 2
  • EP3845922A1 patent drawingFigure 3

AI summary

We disclose sensor systems, and associated calibration systems and methods, that provide efficient and reliable depth and texture fusion. One disclosed method includes transmitting a first light beam from a first perspective, transmitting a second light beam from a second perspective, aligning a visible light photodetector with the second perspective, aligning a depth sensor with the first perspective, and mutually registering the visible light photodetector and the depth sensor using a return of the first light beam, a return of the second light beam, and a reference map. The reference map can include a transform from a reference frame based on the first perspective and a reference frame based on the second perspective.