Camera Array 3D Reconstruction With Real-Time Depth and Velocity

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing 3D scene information generation methods, such as LiDAR and radar, face limitations in angular resolution, computational complexity, and sensitivity to environmental factors, while camera arrays struggle with constrained baselines and high computational demands, making real-time high-resolution 3D scene information challenging.

Innovation Solution

A camera array system comprising multiple cameras capturing images from different viewpoints, with a processing engine that extracts and processes pixel data to generate real-time 3D scene information, using geometric principles and optical flow to determine depth and velocity.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If camera arrays are used to generate 3D scene information, then visual appearance capture is improved, but computational complexity increases

Engineering Contradiction:
Improvevisual appearance captureVSAvoidcomputational complexity
Core Design Contradiction:
Loss of informationVSDevice complexity

Solution Approach 1:

The system performs preliminary actions by pre-positioning multiple cameras with known baselines and pre-extracting pixel data from captured images before computing depth. This preliminary processing organizes the data in advance, reducing the computational burden during real-time 3D reconstruction and enabling efficient triangulation calculations.

Inventive Principle:
Principle #10Preliminary action

2Measurement precision

If LiDAR is used for 3D scene information, then depth measurement is improved, but sensitivity to environmental factors deteriorates

Engineering Contradiction:
Improvedepth measurementVSAvoidsensitivity to environmental factors
Core Design Contradiction:
Measurement precisionVSObject-affected harmful factors

Solution Approach 1:

The patent introduces visual appearance data from multiple cameras as an intermediary to complement depth measurement. By combining optical flow data and pixel information with depth calculations, the system creates a more robust 3D reconstruction that is less sensitive to environmental factors like rain, dust, and snow that affect LiDAR performance.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Measurement precision

If camera arrays with large baseline are used, then depth resolution is improved, but device design difficulty increases

Engineering Contradiction:
Improvedepth resolutionVSAvoiddevice design difficulty
Core Design Contradiction:
Measurement precisionVSEase of manufacture

Solution Approach 1:

The camera array system is designed with universal applicability, where multiple cameras with known baselines can be configured in various arrangements depending on the specific application requirements. The system handles different baseline configurations and camera positions through a unified triangulation algorithm, simplifying the design process while maintaining depth resolution accuracy.

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 accurate, real-time generation of 3D scene information, reducing the need for additional sensors and overcoming environmental interference, facilitating applications like autonomous navigation and security surveillance.

Implementation Method 1

Optical camera systems may be used, with appropriate processing, to generate 3D scene information

Methodology Applied
Scientific EffectLight reflection: Reflection

Implementation Method 2

Binocular disparity methods match local regions in image pairs captured by cameras that have a known physical separation or baseline. From the disparity, a depth for the matched region may be determined based on optical (the assumption that light travels in straight lines) and geometric triangulation principles.

Methodology Applied
Scientific EffectGeometric triangulation: Geometry

Implementation Method 3

A camera array system comprising multiple cameras capturing images from different viewpoints, with a processing engine that extracts and processes pixel data to generate real-time 3D scene information, using geometric principles and optical flow to determine depth and velocity.

Methodology Applied
Scientific EffectOptical flow:

Data Source

PatentEP4139893B1Systems and methods for generating and/or using 3-dimensional information with camera arrays
Publication Date: 2026.02.18 VISIONARY MACHINES PTY LTD
  • EP4139893B1 patent drawingFigure 1
  • EP4139893B1 patent drawingFigure 2
  • EP4139893B1 patent drawingFigure 3

AI summary

The present disclosure is directed to devices, systems and/or methods that may be used for determining scene information from a real-life scene using data obtained at least in part from a camera array. Exemplary systems may be configured to generate three-dimensional information in real-time or substantially real time and may be used to estimate velocity of one or more physical surfaces in a real-life scene.