AR Object Classification for Personalized Media Overlay
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing augmented reality (AR) systems lack the ability to seamlessly integrate computer-generated perceptual information with the real-world environment, failing to provide immersive and interactive experiences due to limitations in object detection and classification.
Innovation Solution
An addressable media system employing computer vision and image processing techniques for accurate object detection and classification, allowing users to associate object classes with user profiles and apply AR media based on these classifications, enabling interactive and immersive AR experiences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If basic object detection is used in AR systems, then the system complexity is low, but the object classification accuracy and AR media integration quality deteriorate
Solution Approach 1:
The system segments the AR media delivery process into distinct phases: object detection, object classification, user profile association, and personalized media application. Each phase is handled by specialized modules that can be independently optimized, allowing high classification accuracy without proportionally increasing overall system complexity.
Solution Approach 2:
The patent introduces intermediary components including an object classification module that bridges detection and media application, a user profile database that mediates between classification and personalization, and a media server that acts as an intermediary between content delivery and AR overlay generation. These intermediaries enable sophisticated functionality while maintaining modular system architecture.
2Adaptability or versatility
If personalized AR media is applied based on user profiles, then the AR experience quality is improved, but the data processing complexity and time consumption increase
Solution Approach 1:
The system performs preliminary actions by pre-establishing user profiles with associated preferences, object class associations, and media preferences before AR sessions begin. The user profile database is pre-populated with classification rules and media content, enabling rapid personalized media application during actual AR sessions without extensive real-time processing.
Solution Approach 2:
The system implements feedback loops where object detection results feed into classification modules, which update user profiles based on detected interactions. These updated profiles then feed back into the media selection process, creating an adaptive system that learns from user behavior patterns and improves personalization over time while optimizing processing efficiency.
Data Source
AI summary
An addressable media system for performing operations that include: accessing image data that depicts an object in an environment at a client device; causing display of a presentation of the image data within a graphical user interface at the client device; detecting the display of the object within the presentation of the image data based on at least a portion of the plurality of image features of the display of the object; identifying an object class based on at least the portion of the image features of the display of the object; receiving an input that selects the display of the object from the client device; and associating the object class that corresponds with the object with the user profile in response to the input that selects the display of the object.