Context-Aware AR Overlay via Learned Object Relationships
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing augmented reality systems lack interactivity between AR images and the real world, resulting in AR images that do not interact intuitively with the real world or the user's context.
Innovation Solution
The system learns object relationships and properties from multivariate content to enhance context-based augmented reality, using techniques such as marker-based tracking, image recognition, and spatial mapping to integrate virtual and physical elements seamlessly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If basic AR overlay systems are used to display computer-generated objects, then the implementation is simple and quick, but the AR images do not interact intuitively with the real world and lack contextual awareness
Solution Approach 1:
The patent introduces machine learning models as intermediary components between the AR display system and the real world. These models process visual input from the physical environment and generate contextual information that mediates the interaction between virtual and real objects, enabling intuitive contextual interactivity without requiring complex manual programming of AR behaviors
Solution Approach 2:
The patent replaces traditional mechanical marker-based tracking systems with machine learning-based visual recognition systems. Instead of requiring predefined markers or anchors in the real world, the system uses ML models to automatically recognize and interpret visual features of physical objects, enabling more natural and versatile AR interactions
2Measurement precision
If marker-based tracking and image recognition are used to recognize anchors in the real world, then the AR images can be positioned on specific locations, but the systems still lack interactivity between AR images and the real world
Solution Approach 1:
The patent transforms static marker-based positioning into dynamic object recognition and interaction. The machine learning models continuously process visual input to identify physical objects and their properties, enabling the AR system to dynamically adapt to different real-world scenarios and enable contextual interactions based on recognized object characteristics
Solution Approach 2:
The patent changes the parameters used for AR positioning and interaction from fixed marker coordinates to dynamic object properties recognized by machine learning models. This includes using object attributes, spatial relationships, and contextual information as parameters for determining AR content placement and interaction behavior
3Loss of information
If sophisticated AR systems with navigation displays are implemented, then graphical instructions can be overlaid, but the systems still do not have interactivity between the AR image and the real world
Solution Approach 1:
The patent implements feedback loops where the machine learning models continuously analyze the real-world environment and adjust AR content accordingly. The system receives feedback from visual recognition of physical objects and their relationships, then modifies AR overlay content to maintain contextual relevance and enable intuitive interactions between virtual and real elements
Data Source
AI summary
A system may generate an AR display in which one or more virtual objects are to be overlaid onto a real world environment, access an Augmented Unification (AU) object comprising one or more properties that define a context of the real world environment based on image recognition performed on the real world environment, identify a virtual object and one or more characteristics of the virtual object based on the AU object, define a behavior of the virtual object with respect to the physical environment based on the one or more context-driven data elements, receive a virtual object to augment the electronic display and one or more permissible actions that can be used based on contextual data, update the electronic display to include the virtual object, and cause an interaction between the physical object and the virtual object based on the one or more permissible actions to be displayed.