Augmented Reality Product Recommendations Through Scene Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing augmented reality (AR) systems require users to manually select and position AR elements, which can interfere with real-world objects and is time-consuming, limiting their functionality and user interest.
Innovation Solution
An AR recommendation system automatically classifies the real-world environment and recommends AR elements to display, removing real-world objects to make room for AR items and intelligently positioning them without user input.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If users manually select and position AR elements, then precise control over AR element placement is achieved, but time consumption increases and user interest decreases
Solution Approach 1:
The system automatically classifies the real-world environment using image processing and machine learning algorithms, and autonomously selects and positions appropriate AR elements without requiring manual user input. The system serves itself by making intelligent decisions about which AR elements to display and where to place them based on environmental analysis.
Solution Approach 2:
The system pre-classifies the real-world environment into categories (indoor, outdoor, urban, rural) and pre-selects appropriate AR elements from a database before the user even requests them. This preliminary preparation eliminates the need for users to manually search and select AR elements during the interaction process.
2Adaptability or versatility
If AR elements are displayed without removing real-world objects, then real-world visibility is preserved, but AR element integration and realism are reduced
Solution Approach 1:
The system extracts or removes the visual representation of real-world objects from the video feed in specific regions, creating transparent or empty spaces where AR elements can be placed. This extraction allows AR elements to be positioned without obscuring the underlying real-world scene, maintaining both realism and visibility.
Solution Approach 2:
The system introduces an intermediary processing layer that separates the real-world video feed from the AR element overlay. By processing the video to identify and remove real-world objects in target regions, the system creates an intermediate state where AR elements can be seamlessly integrated without directly conflicting with real-world visual information.
3Productivity
If automatic environment classification and AR element selection is implemented, then user efficiency is improved, but system complexity increases
Solution Approach 1:
The system segments the complex task of AR element deployment into distinct modules: image processing module for video analysis, environment classification module for categorizing spaces, AR element selection module for choosing appropriate elements, and positioning module for placing elements. This segmentation reduces overall system complexity by making each component independent and manageable.
Solution Approach 2:
The system employs a universal environment classification framework that can handle multiple types of environments (indoor, outdoor, urban, rural) using the same core image processing and machine learning algorithms. This multi-functional approach reduces complexity by avoiding the need for separate specialized systems for each environment type.
Data Source
AI summary
Aspects of the present disclosure involve a system comprising a computer-readable storage medium storing a program and a method for performing operations comprising: receiving a video that includes a depiction of a real-world object in a real-world environment; determining a classification for the real-world environment by processing the real-world object depicted in the video; selecting an augmented reality (AR) item based on the classification of the real-world environment and the real-world object depicted in the video; modifying pixels corresponding to the real-world object depicted in the video to generate a modified video that excludes the depiction of the real-world object; and adding the AR item to the modified video at a display position corresponding to the modified pixels.