AR Narrative System Using Object Recognition for Indoor Location
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing Augmented Reality (AR) systems struggle to provide interactive and location-specific narratives in indoor environments, where GPS signals are unreliable.
Innovation Solution
The development of an AR system that uses AI-powered image classifiers to recognize physical objects and determine their attributes, allowing for the creation of interactive narratives that adapt based on the user's location and actions within the physical environment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If GPS-based location tracking is used for AR narratives, then outdoor location accuracy is improved, but indoor location reliability deteriorates due to signal penetration issues
Solution Approach 1:
The patent introduces image classifiers as intermediary components that mediate between the physical environment and the AR narrative system. Instead of relying directly on GPS signals, the system uses image recognition of physical objects as an intermediate step to determine location and trigger appropriate narratives, solving the indoor signal penetration problem.
Solution Approach 2:
The patent replaces the GPS satellite-based electromagnetic signal system with an AI-based image recognition system. Instead of using mechanical/electromagnetic signal transmission for location tracking, the system uses optical image capture followed by AI classification to determine physical location, enabling reliable indoor operation.
2Adaptability or versatility
If AI image classifiers are integrated into AR systems for object recognition, then narrative adaptability to physical environment is improved, but device complexity increases
Solution Approach 1:
The patent makes the image classifier a universal component that serves multiple functions: identifying physical objects, determining user location, triggering narratives, and adjusting narrative content. This multi-functionality reduces the need for separate specialized components, managing complexity while enhancing adaptability.
Solution Approach 2:
The AR system uses the image classifier to automatically determine location and select appropriate narratives without requiring manual user input or complex configuration. The system serves itself by autonomously adapting the narrative based on what it detects in the physical environment, reducing operational complexity.
3Measurement precision
If multiple image classifiers are used to recognize different physical objects, then object recognition accuracy is improved, but processing time increases
Solution Approach 1:
The patent segments the object recognition task by using specialized image classifiers for different types of physical objects (e.g., one classifier for furniture, another for electronics). Each classifier is optimized for its specific object type, improving accuracy while allowing parallel processing to reduce overall processing time.
Solution Approach 2:
The system implements partial object recognition by using multiple classifiers that process different aspects or categories of objects simultaneously. Rather than waiting for complete recognition of all objects, the system can trigger narratives based on partial recognition results, reducing processing time while maintaining adequate accuracy.
Data Source
AI summary
Systems and methods of using augmented reality (AR) with object recognition (OR) in a physical environment to advance a narrative may be provided. For example, the system may access narrative data representing the narrative. The system may, for a first node of the one or more nodes: generate a user interface associated with the narrative for the first node and access a first decision rule specifying a first physical object to be recognized to advance past the first node in the narrative. The system may further perform image recognition on an image of the physical environment, determine that the first physical object is in the physical environment based on the image recognition, transition from the first node in the narrative based on the first decision rule and the determination, and update the user interface to a second node in the narrative based on the transition from the first node.