Viewer Engagement Measurement Using Camera-Based Face Orientation Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for measuring TV audience engagement are flawed as they only count the number of people in the room, do not gauge actual viewing habits, and fail to assess reactions to programs or advertisements, lacking demographic specificity and accuracy in ratings.
Innovation Solution
A system that uses cameras and microphones to capture image and audio data from a viewing area, processing this data locally to determine the number of people and their engagement with the content, including facial recognition and sentiment analysis, and transmitting processed data to a remote server for further analysis, allowing for the calculation of viewability and attention indices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional people meters and diaries are used to collect audience data, then the number of people in the room can be counted, but the actual viewing habits and engagement levels cannot be measured
Solution Approach 1:
The patent replaces conventional mechanical/people-meter-based audience measurement systems with computer vision technology using cameras and image processing algorithms. The system captures images of the viewing area, detects faces, determines facial orientations, and analyzes eye positions to automatically assess viewer engagement without requiring manual input from viewers.
Solution Approach 2:
The patent introduces an intermediary processing system that acts between the viewer and the measurement system. This intermediary includes image capture devices, image processing modules, and analysis algorithms that translate physical viewer behavior (face orientation, eye position) into quantifiable engagement metrics, bridging the gap between passive viewing and measurable data.
2Measurement precision
If TV ratings are aggregated at household level, then overall audience size can be measured, but demographic-specific engagement data is lost
Solution Approach 1:
The patent segments the audience measurement data by demographic characteristics. The image processing system not only counts viewers but also identifies demographic attributes (such as age groups, gender) and creates separate engagement metrics for different demographic segments. This allows advertisers and content providers to target specific demographics with precision while maintaining a manageable data structure.
3Measurement precision
If simple presence detection is used, then the system remains simple to operate, but it cannot distinguish between people in the room and actual viewers
Solution Approach 1:
The patent implements a self-service measurement system that automatically performs all detection and analysis functions without requiring user configuration or input. The camera system automatically captures images, the processing algorithms automatically detect faces and analyze orientations, and the system automatically generates engagement metrics. This maintains ease of operation while achieving high measurement precision.
Data Source
AI summary
A system and method for quantifying viewer engagement with a video playing on a display in a respondent household includes an agreed upon camera arrangement to monitor viewer engagement. The system and method includes the ability to determine what sources of content are being accessed by the household, and other data such as time of viewing, and source of content.