Hybrid Presence Detection via Meter Data Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing audience monitoring systems face challenges in accurately determining the presence status of audience members, particularly due to issues with audience compliance and the processor intensity of facial recognition in passive people meters.
Innovation Solution
The proposed solution combines a passive people meter for capturing audience images and an active people meter for prompting audience members to verify their presence, using a combination of facial recognition and active prompting to generate presence status data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If facial recognition is used in passive people meters to determine audience presence, then measurement precision is improved, but device complexity and processing requirements increase
Solution Approach 1:
The patent combines passive people meter technology (facial recognition cameras) with active people meter technology (portable devices with prompts) into a hybrid system. The passive component captures images for automated recognition, while the active component provides verification prompts to panelists, merging both approaches to achieve accurate presence detection without over-relying on computationally intensive facial recognition alone.
Solution Approach 2:
The portable active people meter acts as an intermediary between the passive camera system and the central facility. It captures verification data from panelists via prompts and transmits this information to corroborate or refute facial recognition results, serving as a mediator that reduces the burden on the passive system while maintaining measurement accuracy.
2Ease of operation
If only passive people meters with facial recognition are used, then ease of operation is improved, but reliability of presence status detection deteriorates due to audience compliance issues
Solution Approach 1:
The system implements feedback loops where the passive people meter continuously monitors the media environment and compares detected presence against expected panelist presence. When discrepancies are detected (e.g., someone should be present but isn't detected, or vice versa), the system generates prompts for verification, creating a feedback mechanism that maintains reliability while keeping the system largely automated.
Solution Approach 2:
Instead of requiring continuous active verification from all panelists, the system uses partial action by only prompting verification when the passive facial recognition system detects potential presence status changes or discrepancies. This selective verification approach maintains reliability without requiring excessive active participation from all panelists at all times.
Data Source
AI summary
Methods, apparatus, systems, and articles of manufacture are disclosed to detect a presence status. An example apparatus includes media identification circuitry to generate first signatures representative of first audio data associated with a monitored media device, a comparator to obtain second signatures from a portable meter, the second signatures representative of second audio data sensed by the portable meter and compare the first signatures and the second signatures to determine a comparison result, presence detection circuitry to determine a presence status of a user based on the comparison result, the user associated with the portable meter, and network communication circuitry to transmit the presence status to a data processor to perform audience measurement based on the presence status.