Crowd-Sourced AR Venue Mapping Using Distributed End-User Devices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional inventory management systems in retail environments struggle to provide real-time, accurate mapping and localization of products and inventory due to limitations in camera views and robotic device capabilities, leading to inefficiencies in restocking, customer experience, and inventory insights.
Innovation Solution
A system utilizing crowd-sourced images and videos from end-user devices for real-time or near-real-time mapping and localization, combining Wi-Fi and computer vision data with Machine Learning to create an up-to-date planogram of store inventory, enabling precise product location and demand analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional inventory management systems use fixed camera views and robotic devices for mapping, then the system structure is simple, but the real-time mapping accuracy and coverage are insufficient
Solution Approach 1:
The patent segments the mapping task by using multiple end-user devices distributed throughout the store instead of a single complex robotic system. Each device independently captures local images, and the system aggregates these segments to create a comprehensive real-time map, resolving the contradiction between accuracy and complexity.
Solution Approach 2:
The system leverages end-user devices (smartphones, tablets) that customers already have to perform the mapping function. Instead of deploying specialized robotic devices, the patent enables ordinary customers to contribute to inventory mapping through their existing devices, eliminating the need for complex dedicated infrastructure.
2Speed
If the system uses limited robotic devices for inventory mapping, then the device complexity is low, but the speed of mapping and localization is slow
Solution Approach 1:
The patent merges the mapping capabilities of numerous end-user devices into a single coordinated system. By combining data from multiple simultaneous captures across different locations, the system achieves rapid comprehensive mapping much faster than any single robotic device could provide.
Solution Approach 2:
The system continuously updates the inventory map by aggregating data streams from multiple end-user devices that are constantly moving and capturing images throughout the store. This continuous action enables real-time mapping without the delays associated with sequential robotic scanning.
3Measurement precision
If ceiling cameras are used for store mapping, then the installation is simple, but the field of view is limited and cannot capture true details
Solution Approach 1:
The patent transitions from a top-down ceiling camera perspective to a ground-level customer perspective. By positioning cameras at eye level with customers holding their devices, the system captures detailed product information and spatial relationships that ceiling cameras cannot resolve, adding a new dimensional perspective to the mapping.
Solution Approach 2:
Instead of relying on ceiling cameras to capture everything from above, the system uses multiple local views from end-user devices positioned throughout the store. Each device provides high-quality local imaging of specific product areas, and the system aggregates these localized details to achieve comprehensive accurate mapping.
4Loss of information
If the system provides detailed real-time inventory mapping, then the inventory management quality is high, but the data processing complexity and computational requirements increase
Solution Approach 1:
The patent extracts and processes only the essential information needed for inventory mapping from the vast amount of image data captured by end-user devices. By focusing computation on identifying products, their locations, and quantities rather than processing every pixel, the system achieves high inventory accuracy without excessive computational complexity.
Solution Approach 2:
The system processes a subset of the most informative data points from each image (such as detected product locations and identifiers) rather than analyzing every detail. This partial processing approach provides sufficient inventory mapping accuracy while significantly reducing the computational burden compared to complete image analysis.
Data Source
AI summary
System, device, and method of Augmented Reality based mapping of a venue and navigation within a venue. A method includes: performing a crowd-sourced mapping process, that maps a retail store and maps particular products sold within that retail store, based on computer-vision analysis of a plurality of images captured by a plurality of end-user devices of customers within that retail store; and generating a representation of a store map reflecting actual real-time location of particular products within that retail store. Turn-by-turn walking directions are provided, to guide the user from his current in-store location towards a destination product within that retail store. Augmented Reality promotions, advertisements and marketing content elements, route guidance, and other content are generated and displayed on the end-user device.