Display Interface Facial Recognition for Average-Person Presentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Smart TVs lack the ability to recognize and provide information on average persons in a scene image, limiting the range of content recognition and presentation capabilities.
Innovation Solution
A display apparatus and method that includes a controller to receive user commands, obtain facial feature information, match it with a database, and generate sharing controls for average-person recognition data, enabling presentation of additional information on individuals in a scene image.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the smart TV uses traditional screenshot recognition to identify persons, then it can recognize celebrities with high confidence, but it cannot recognize average persons whose facial features do not match the database
Solution Approach 1:
The patent introduces an intermediary mechanism (similar facial feature information retrieval) between the facial feature extraction and the database matching. When no exact match is found, the system retrieves similar facial features and generates average-person recognition data, thereby mediating between the binary match/fail outcome and providing nuanced recognition results for average persons.
Solution Approach 2:
The patent changes the recognition parameter from binary matching (match/no match) to a gradient approach by introducing similarity degrees and confidence levels. By retrieving similar facial features and calculating matching confidence levels, the system transforms the rigid recognition parameter into a flexible one that can accommodate both celebrities and average persons.
2Loss of information
If the system retrieves similar facial feature information for average persons, then it can provide more comprehensive person information, but it increases the complexity of the recognition process
Solution Approach 1:
The patent applies preliminary action by pre-storing facial feature information in a database and preparing the retrieval mechanism in advance. When recognition is needed, the system can quickly query and retrieve similar facial features without performing complex real-time analysis, thereby reducing the actual recognition process complexity while maintaining information completeness.
Solution Approach 2:
The patent uses copying by retrieving similar facial feature information from the database instead of performing complex real-time facial analysis. The system copies pre-processed facial feature data that matches the input within a confidence threshold, simplifying the recognition process while still providing comprehensive person information.
3Loss of information
If the smart TV presents detailed recognition information on the display, then it enriches user experience, but it increases the amount of data processing and presentation required
Solution Approach 1:
The patent applies local quality by presenting different levels of recognition information in different display regions or contexts. The system can show detailed information for recognized persons while maintaining the original content elsewhere, providing enriched user experience without overwhelming the entire display interface and maintaining processing efficiency.
Solution Approach 2:
The patent uses partial action by selectively presenting recognition information based on user needs and display context. Instead of processing and displaying all possible person information uniformly, the system presents relevant information partially or selectively, enriching user experience where needed while maintaining data processing efficiency overall.
Data Source
AI summary
Provided are a display apparatus and a person recognition and presentation method. The display apparatus includes a display and a controller that is in communication with the display. The controller is configured to: associated information of a display interface of the display and generate a scenario image for recognition in response to a user command; obtain facial feature information for recognition in the scenario image; obtain similar facial feature information when a matching confidence level of pre-stored facial feature information in a database with the facial feature information for recognition does not exceed a preset confidence level; obtain average-person recognition data; generate a sharing control uniquely matching with the facial feature information for recognition; and control the display to present the average-person recognition data and the sharing control on a current display interface.