Marker-Free AR Image Synthesis via Feature Point Displacement

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing augmented reality systems require a marker to be present in the image pickup area for accurate synthesis of additional images with the real environment, limiting their functionality when the marker is not included.

Innovation Solution

An information processing apparatus and method that uses a first image pickup means, a positional relationship estimating unit, and an image generating unit to synthesize additional images with the real environment, even without a marker, by estimating the positional relationship based on a registered marker image or feature point displacement, allowing for the generation and display of synthetic images.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a marker is required in the image pickup area for accurate positional relationship estimation, then the synthesis accuracy of additional images is improved, but the versatility and usability of the system deteriorates when the marker is not included

Engineering Contradiction:
Improvepositional relationship estimation accuracyVSAvoidsystem functionality without marker
Core Design Contradiction:
Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The patent introduces feature points as an intermediary element between the marker-based system and the environment. When markers are not present, the system extracts feature points from the background image and uses their displacement to estimate positional relationships, thereby mediating the transition from marker-dependent to marker-independent operation

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The system dynamically changes the parameter used for positional relationship estimation based on the presence or absence of markers. When markers are detected, it uses marker-based parameters; when markers are absent, it switches to feature point displacement parameters, thereby adapting to different environmental conditions

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If the system relies on marker detection for additional image display, then the positional accuracy is improved, but the operational flexibility deteriorates

Engineering Contradiction:
Improveadditional image positioning accuracyVSAvoidoperation without marker constraints
Core Design Contradiction:
Measurement precisionVSEase of operation

Solution Approach 1:

The patent implements a dynamic system that automatically adjusts its operation mode based on marker detection results. The system transitions between marker-based mode (high precision) and feature point-based mode (flexible operation), allowing it to maintain ease of operation across different scenarios while preserving positioning accuracy when possible

Inventive Principle:
Principle #15Dynamics

3Reliability

If the image pickup area must include a marker for synthesis, then the synthesis reliability is improved, but the application scope deteriorates

Engineering Contradiction:
Improveimage synthesis reliabilityVSAvoidapplication environment range
Core Design Contradiction:
ReliabilityVSAdaptability or versatility

Solution Approach 1:

The system performs preliminary feature point extraction and registration during the marker detection phase. By pre-processing and storing feature point information when markers are present, it prepares the necessary data structures and parameters that enable reliable synthesis operation even when markers are subsequently absent, thereby extending application scope

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS8749583B2Information processing apparatus, information processing system, and information processing method
Publication Date: 2014.06.10 SONY GROUP CORP
  • US8749583B2 patent drawing
  • US8749583B2 patent drawing
  • US8749583B2 patent drawing

AI summary

A first image pickup means picks up an image of a real space, to thereby obtain a first background image.A first positional relationship estimating unit estimates, in a case where the first background image includes a first marker image being an image of a marker previously registered and located in the real space, a first positional relationship being a spatial positional relationship of the first image pickup means to the marker based on the first marker image, and estimates, in a case where the first background image fails to include the first marker image, the first positional relationship based on a displacement of a feature point extracted from the first background image.An image generating unit generates a first additional image based on the first positional relationship, and synthesizes the first additional image and the first background image, to thereby generate a synthetic image.