Multi-Baseline Camera Array for Depth Augmentation in VR/AR

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current technologies face challenges in accurately determining depth information for both near-field and far-field objects in virtual reality/augmented reality applications, particularly in achieving precise depth estimation across a wide operating range and varying illumination conditions.

Innovation Solution

A multi-baseline camera array system is employed, which includes a set of cameras with different baseline distances and imaging properties. This system generates an initial depth map, identifies near-field and far-field portions of the scene, and refines the depth map using image data from near-field and far-field cameras, respectively. Additionally, the system uses an illumination light source positioned close to a reference camera to enhance depth estimation accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a single camera system is used, then device complexity is reduced, but depth estimation precision deteriorates across varying distances

Engineering Contradiction:
Improvedepth estimation precisionVSAvoidcamera system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The camera system is segmented into multiple camera modules with different baseline distances. Near-field cameras capture close objects with high precision, while far-field cameras capture distant objects. This segmentation allows each camera subset to be optimized for specific depth ranges, resolving the contradiction between measurement precision and device complexity by dividing the monolithic system into functional segments.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system dynamically selects and switches between different camera subsets based on the detected distance to objects. When near-field objects are detected, the system activates near-field cameras; when far-field objects are detected, it switches to far-field cameras. This dynamic adaptation maintains high depth estimation precision across varying distances without requiring all cameras to operate simultaneously, thus managing device complexity.

Inventive Principle:
Principle #15Dynamics

2Measurement precision

If multiple camera subsets are used for different depth ranges, then depth estimation precision is improved, but device complexity increases

Engineering Contradiction:
Improvedepth map accuracyVSAvoidcamera array configuration
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

Each camera subset is designed to serve multiple functions: near-field cameras can capture both color and depth information for close objects, while far-field cameras do the same for distant objects. This multi-functionality reduces the need for entirely separate systems for different depth ranges, as each camera subset can independently handle multiple imaging tasks, thereby improving depth map accuracy without proportionally increasing device complexity.

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

3Measurement precision

If cameras with different baseline distances are deployed, then depth estimation precision across wide range is improved, but system configuration complexity increases

Engineering Contradiction:
Improvedepth precision across distance rangeVSAvoidcamera configuration
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

Different camera subsets are configured with locally optimized baseline distances suited for their specific operational ranges. Near-field cameras use smaller baselines appropriate for close objects, while far-field cameras use larger baselines suitable for distant objects. This local optimization of camera parameters to match local operational requirements improves depth precision across the entire distance range without requiring a uniformly complex system configuration.

Inventive Principle:
Principle #3Local quality

4Measurement precision

If illumination sources are positioned close to reference cameras, then depth estimation accuracy is improved, but system complexity increases

Engineering Contradiction:
Improvedepth estimation accuracyVSAvoidillumination system configuration
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The illumination sources are merged with the reference camera assemblies, positioning them in close proximity to the reference cameras. This merging allows the illumination system to be integrated with the existing camera structure, improving depth estimation accuracy through coordinated lighting while avoiding the need for completely separate illumination subsystems, thus limiting the increase in overall system complexity.

Inventive Principle:
Principle #5Merging (Combining)

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

The multi-baseline camera array system effectively estimates depth with high precision across a wide range of distances and varying illumination conditions, improving the accuracy of virtual object placement and interaction with real-world objects in AR/VR applications.

Implementation Method 1

Systems and methods for estimating depth with camera arrays... generating an initial depth map of a scene... refine the depth map for the near-field portions of the scene using image data captured from a near-field set of cameras

Methodology Applied
Scientific EffectParallax: Parallax

Data Source

PatentUS20250142038A1Multi-Baseline Camera Array System Architectures for Depth Augmentation in VR/AR Applications
Publication Date: 2025.05.01 ADEIA IMAGING LLC
  • US20250142038A1 patent drawing
  • US20250142038A1 patent drawing
  • US20250142038A1 patent drawing

AI summary

Embodiments of the invention provide a camera array imaging architecture that computes depth maps for objects within a scene captured by the cameras, and use a near-field sub-array of cameras to compute depth to near-field objects and a far-field sub-array of cameras to compute depth to far-field objects. In particular, a baseline distance between cameras in the near-field subarray is less than a baseline distance between cameras in the far-field sub-array in order to increase the accuracy of the depth map. Some embodiments provide an illumination near-IR light source for use in computing depth maps.