Automatic AR Marker Registration via Image Feature Extraction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The existing augmented reality (AR) systems require manual registration of AR markers in a global coordinate system, which is cumbersome and necessitates positioning markers on walls or similar surfaces, limiting their flexibility and efficiency.

Innovation Solution

An information processing apparatus with a camera, memory, and processor that automatically extracts feature values from images to determine location, allowing for the use of any image with a high feature value score as a marker, eliminating the need for manual marker placement and registration.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If manual registration of AR markers in global coordinate system is used, then tracking accuracy is achieved, but operation complexity and setup time increase significantly

Engineering Contradiction:
Improvetracking accuracyVSAvoidmarker registration complexity
Core Design Contradiction:
Measurement precisionVSEase of operation

Solution Approach 1:

The system performs automatic marker detection and coordinate system establishment without requiring manual intervention. The AR device automatically detects markers in the environment, extracts their feature values, and establishes the relationship between marker coordinate systems and the global coordinate system autonomously, eliminating the need for manual coordinate input and marker positioning.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent replaces the manual mechanical process of positioning markers on walls and inputting coordinates with an automated image processing system. The system uses camera imaging, feature value extraction, and automatic coordinate calculation to substitute the manual mechanical operations, thereby reducing operation complexity while maintaining tracking accuracy.

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

2Reliability

If traditional AR markers are positioned on walls, then stable tracking is achieved, but system flexibility and adaptability decrease

Engineering Contradiction:
Improvetracking stabilityVSAvoidmarker placement flexibility
Core Design Contradiction:
ReliabilityVSAdaptability or versatility

Solution Approach 1:

The system enables any image-containing object to function as a marker, not just traditional wall-mounted markers. By extracting feature values from images captured by the camera, the system allows diverse objects such as posters, product packaging, and other image-bearing surfaces to serve as tracking markers, greatly enhancing system versatility while maintaining tracking stability through reliable feature value matching.

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

3Manufacturing precision

If manual marker input and registration is required, then precise marker location control is achieved, but productivity and setup efficiency decrease

Engineering Contradiction:
Improvemarker location controlVSAvoidsetup efficiency
Core Design Contradiction:
Manufacturing precisionVSProductivity

Solution Approach 1:

The system pre-establishes a database of marker feature values and their corresponding locations. During operation, the device automatically compares captured image feature values against this pre-stored database to quickly identify and track markers, eliminating the need for manual coordinate input and significantly improving setup efficiency while maintaining precise location control through the pre-configured marker information.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS11321933B2Information processing apparatus configured to determine a location of a candidate for a marker for augmented reality
Publication Date: 2022.05.03 TOSHIBA TEC KK
  • US11321933B2 patent drawing
  • US11321933B2 patent drawing
  • US11321933B2 patent drawing

AI summary

An information processing apparatus includes a camera acquiring an image, a memory storing a feature value of an image associated with a location, and a processor. When a current location of the apparatus is not stored in the memory, the processor extracts a first area from an acquired image, acquires a first feature value of the first area, and when a feature value that is substantially identical with the first feature value is stored in the memory, determines the current location using a location stored in the memory in association with the feature value. When the current location is stored in the memory, the processor extracts a second area from an acquired image, acquires a second feature value of the second area, determines a location corresponding to the second area using the current location, and stores in the memory the determined location in association with the second feature value.