AR Location Fingerprinting via Visual Acoustic Analysis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current augmented reality systems face challenges in accurately determining real-world locations, particularly indoors, due to the inaccuracy of location data from location-aware devices, which affects the precise placement of virtual objects within real-world scenes.

Innovation Solution

The system employs computer vision techniques to detect and analyze real-world features such as walls, surfaces, and objects, using methods like edge detection and machine learning algorithms (e.g., Faster R-CNN, Mask R-CNN) to generate a 3D fingerprint of a location, allowing for precise positioning of virtual objects relative to these features, and utilizes audio analysis for further accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If location data from location-aware devices is used for virtual object placement, then the system is simple to operate, but the placement precision deteriorates due to inaccuracy of location data

Engineering Contradiction:
Improveease of operationVSAvoidplacement precision
Core Design Contradiction:
Ease of operationVSManufacturing precision

Solution Approach 1:

The patent introduces computer vision techniques and audio analysis as intermediary systems between the location-aware device and the virtual object placement. These intermediaries capture visual and acoustic data of the real-world environment to create a fingerprint that serves as a mediator for determining device coordinates, thereby improving placement precision without complicating the user operation

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent replaces the mechanical/location-based positioning system with an optical and acoustic sensing system. Instead of relying solely on location data from GPS or other location-aware mechanisms, the system substitutes these with camera-based computer vision and microphone-based audio analysis to determine device coordinates and improve placement accuracy

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Manufacturing precision

If computer vision techniques and audio analysis are employed to generate location fingerprint, then the placement precision is improved, but the device complexity increases

Engineering Contradiction:
Improveplacement precisionVSAvoiddevice complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent makes the client device multi-functional by enabling it to perform not only location-based operations but also computer vision processing and audio analysis. The device's camera and microphone, originally for general purposes, are repurposed to capture environmental data for location fingerprinting, thereby improving precision without requiring additional specialized hardware

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

Solution Approach 2:

The system enables the client device to self-determine its coordinates by capturing and analyzing its own visual and acoustic environment. The device uses its built-in sensors (camera and microphone) to generate the location fingerprint independently, reducing reliance on external positioning infrastructure and minimizing the need for additional complex components

Inventive Principle:
Principle #25Self-service

Data Source

PatentUS11398088B2Systems, methods and apparatuses to generate a fingerprint of a physical location for placement of virtual objects
Publication Date: 2022.07.26 ADEIA MEDIA HOLDINGS INC
  • US11398088B2 patent drawing
  • US11398088B2 patent drawing
  • US11398088B2 patent drawing

AI summary

Systems, methods and apparatuses to generate a Fingerprint of a Physical Location for Placement of Virtual Objects are disclosed. In one aspect, embodiments of the present disclosure include a method, which may be implemented on a system, to generate a fingerprint of the physical location using the location metadata including one or more of, the geometric features and the physical objects. The method can further include tracking device movement in the physical location to determine a position of the device relative to the fingerprint of the physical location and rendering or placing the virtual object in the physical location based on the position of the device and the fingerprint of the physical location.