Face Recognition System for TV Viewer Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing face recognition technologies for TV viewer identification are computationally intensive and unsuitable for limited CPU devices, making it difficult to provide personalized TV experiences for groups, as they struggle to efficiently process multiple individuals and often fail to accurately identify the primary viewer holding the remote control.
Innovation Solution
A method that enrolls users in a database with biometric data, uses a camera to scan and isolate faces, matches extracted biometric data with stored profiles, and updates profile IDs based on repeated detections, with optional optimizations such as reduced scanning areas and prioritizing likely viewer positions, and utilizing a remote control beacon light to identify the primary user.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If face recognition algorithm is run on wide image encompassing multiple people, then all TV viewers can be identified for personalized experience, but CPU load increases proportionally to number of people in scene
Solution Approach 1:
The wide image is divided into multiple sub-regions or zones, and face recognition is performed independently in each region. This segmentation reduces the computational complexity by limiting the search space in each processing unit while still covering the entire scene with multiple people.
Solution Approach 2:
The problem is shifted from 2D image processing to 3D spatial processing by utilizing depth information or multi-camera arrays. This allows the system to track multiple faces simultaneously in different spatial zones, reducing the computational burden on a single CPU by distributing processing across multiple dimensions.
2Device complexity
If face recognition algorithm is optimized for single user, then CPU load is reduced, but ability to provide personalized experience for group of people deteriorates
Solution Approach 1:
User profiles and preferences are pre-configured and stored in the system before actual viewing. When multiple people are detected, the system quickly matches detected faces against pre-stored profiles and immediately assigns the corresponding personalized experience, avoiding real-time complex processing for each user.
Solution Approach 2:
The system creates simplified representations or models of each user's viewing preferences and behavior patterns. These copied profiles are then used for quick matching and personalization decisions, reducing the need for complex real-time analysis of multiple users simultaneously.
3Measurement precision
If standard face recognition is used to identify primary viewer holding remote control, then accuracy may be insufficient, but implementing advanced tracking increases device complexity
Solution Approach 1:
The remote control itself serves as an intermediary object to identify the primary viewer. The system detects the remote control's position and orientation in the scene, and the person holding it is automatically identified as the primary viewer. This intermediary approach provides high accuracy without requiring complex facial recognition algorithms.
Solution Approach 2:
The system replaces complex optical-mechanical face recognition systems with simpler sensor-based detection methods. By using motion sensors, position sensors, or simple image analysis to detect the remote control holder, the system achieves accurate primary viewer identification with minimal device complexity.
Data Source
AI summary
Method to customize an application associated with a television experience based on the recognition of users located in front of a display and in the field of view of a camera, comprising the following steps:an initialization step during which each user is enrolled in a database of a computer system and is defined by a profile referenced by a profile ID and comprising the user name, biometric data and additional personal data,a finding step during which a wide image, acquired by said camera is scanned to isolate at least one user's faces, to define a marking area surrounding it, to memorize the position of said marking areas,a matching step to extract the biometric data from said marking area, to match them with the biometric data of the profiles stored in the database, and to assign the detected profile ID with the marking area.While subsequent identification is requested by the application, the following steps are executedacquiring a wide image with the camera,extracting the areas of said image according to the marking areas,extracting for a particular marking area the biometric data of a face,starting the comparison of the extracted biometric data with the biometric data of the profile ID related to this marking area, and in absence of match, continuing with the other biometric data of the database until one profile is found,transmitting the found profile ID to the application.