Indoor Image Localization Using Angular Descriptors and Floor Plans

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods face challenges in accurately determining the acquisition location of images within building interiors without the use of depth sensors or distance-measuring devices, and in effectively representing and utilizing building interior information for navigation and visualization.

Innovation Solution

The system automatically determines the acquisition location of images by analyzing visual content, using image angular descriptors to match with building angular descriptors, and integrates this information with floor plans to enhance navigation and visualization without requiring depth sensors or distance-measuring devices.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If depth sensors or distance-measuring devices are used to determine image acquisition locations, then measurement precision is improved, but device complexity and cost increase

Engineering Contradiction:
Improveimage acquisition location determination accuracyVSAvoidsensor system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent replaces depth sensors and distance-measuring devices with a computational approach using 2D image analysis. The system uses angular descriptors extracted from standard 2D images to determine acquisition locations, substituting mechanical/optical measurement systems with image processing algorithms. This eliminates the need for additional sensors while maintaining location determination accuracy.

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

Solution Approach 2:

The patent creates angular descriptors as computational representations (copies) of the physical spatial relationships in the environment. Instead of directly measuring distances with sensors, the system captures angular relationships from 2D images and uses these angular copies to infer acquisition locations, achieving the same goal without physical distance-measuring devices.

Inventive Principle:
Principle #26Copying

2Manufacturing precision

If floor plans are manually constructed and maintained to represent building interiors, then visualization accuracy is improved, but ease of manufacture and updates deteriorate

Engineering Contradiction:
Improvefloor plan accuracyVSAvoidfloor plan construction and update difficulty
Core Design Contradiction:
Manufacturing precisionVSEase of manufacture

Solution Approach 1:

The patent enables floor plans to update themselves automatically by processing new images as they are captured. When a new image is acquired, the system extracts angular descriptors, compares them with existing floor plan data, and automatically updates the floor plan without manual intervention. This self-updating mechanism eliminates the need for manual floor plan maintenance while preserving accuracy.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent performs preliminary actions by continuously capturing images and pre-processing them to extract angular descriptors, which are then ready for immediate comparison and integration into the floor plan. This preliminary preparation of data ensures that when updates are needed, they can be executed quickly and accurately without manual floor plan reconstruction.

Inventive Principle:
Principle #10Preliminary action

3Loss of information

If comprehensive building interior information is captured and stored for remote viewing, then information completeness is improved, but data management complexity and storage requirements increase

Engineering Contradiction:
Improvebuilding interior information completenessVSAvoiddata management complexity
Core Design Contradiction:
Loss of informationVSDevice complexity

Solution Approach 1:

The patent extracts only the essential angular descriptor information from complete building interior images. Instead of storing and transmitting entire high-resolution images or all possible interior details, the system extracts and stores only the angular relationships and spatial configuration data needed for floor plan reconstruction and navigation. This selective extraction maintains information completeness for the intended purpose while dramatically reducing data management complexity.

Inventive Principle:
Principle #2Taking out (Extraction)

Data Source

PatentEP3965061A1Automated analysis of image contents to determine the acquisition location of the image
Publication Date: 2022.03.09 MFTB HOLDCO INC
  • EP3965061A1 patent drawingFigure 1A~1B
  • EP3965061A1 patent drawingFigure 2A
  • EP3965061A1 patent drawingFigure 2B

AI summary

Techniques are described for using computing devices to perform automated operations for determining the acquisition location of an image using an analysis of the image's visual contents. In at least some situations, images to be analyzed include panorama images acquired at acquisition locations in an interior of a multi-room building, and the determined acquisition location information includes a location on a floor plan of the building and in some cases orientation direction information - in at least some such situations, the acquisition location determination is performed without having or using information from any distance-measuring devices about distances from an image's acquisition location to objects in the surrounding building. The acquisition location information may be used in various automated manners, including for controlling navigation of devices (e.g., autonomous vehicles), for display on one or more client devices in corresponding graphical user interfaces, etc.