Smart TV Ad Fingerprinting for Accurate Audience Measurement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current audience measurement systems for television commercials are inaccurate, particularly for non-traditional media and advertisements, as they fail to verify if the television is on and what content is being viewed, especially with the rise of on-demand and over-the-top programming, and are not equipped to handle the increasing complexity of advertisement distribution.
Innovation Solution
A system and method utilizing an iSpot Server and Smart TV devices to collect and analyze data on television viewing habits, including advertisements, through fingerprinting and watermark recognition, to accurately measure audience exposure and View Rates across diverse platforms, including linear TV, on-demand, and OTT services.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional audience measurement systems (People Meters, paper logs) are used, then implementation is straightforward, but measurement precision deteriorates due to inability to verify actual viewing and content accuracy
Solution Approach 1:
The patent introduces watermarks as an intermediary element embedded in advertisements and content. These watermarks serve as mediators between the content delivery system and the measurement system, enabling automatic detection and verification of what is actually being displayed without requiring complex active participation from viewers or sophisticated inference algorithms
Solution Approach 2:
The patent replaces manual mechanical systems (People Meters requiring button pressing, paper logs requiring handwriting) with automated electronic detection systems that use fingerprinting and watermark recognition technology to automatically measure and verify content viewing without human intervention
2Ease of operation
If Set-Top Box data is used for measurement, then ease of operation improves due to existing infrastructure, but measurement precision deteriorates due to inability to verify actual viewing and advertisement content
Solution Approach 1:
The patent applies preliminary action by embedding watermarks and fingerprints in advertisements and content before they are delivered to Set-Top Boxes. This pre-prepared measurement infrastructure allows the existing Set-Top Box infrastructure to be leveraged for accurate measurement without requiring modifications to user behavior or additional active components
Solution Approach 2:
The patent implements feedback mechanisms where the detection system continuously monitors and compares actual displayed content against the expected content schedule. When discrepancies are detected (such as local insertions or deviations from schedule), the system automatically adjusts measurements to reflect what was actually viewed rather than what was scheduled
3Measurement precision
If comprehensive advertisement tracking is implemented across all platforms, then measurement precision improves, but device complexity increases due to handling diverse formats and insertion methods
Solution Approach 1:
The patent applies universality by designing a single watermark-based measurement system that can detect and track advertisements across multiple platforms and formats (broadcast TV, cable, satellite, OTT, local insertions). The fingerprinting and watermark recognition technology serves multiple functions: identifying content, verifying advertisements, detecting local insertions, and measuring viewing across diverse delivery mechanisms without requiring separate systems for each platform
Solution Approach 2:
The patent uses parameter changes by adapting the measurement approach based on the delivery platform and content type. The system adjusts detection parameters, watermark placement, and verification methods according to the specific characteristics of each platform (broadcast vs. on-demand vs. OTT) and content format, enabling accurate tracking across diverse environments through a unified flexible framework
Data Source
AI summary
Television transmissions are received at multiple locations across a large area. Advertisements may be identified, harvested and tagged from within in the television transmissions. The advertisements may be assigned identifiers; media plans may be determined. Advertisement representations (e.g. fingerprints) may be used to identify advertisements in television content received by the smart TVs. The smart TVs may report the fingerprints along with other identifiers or samples thereof. Television content and advertisements therein as rendered by the smart TVs may be categorized as live, timeshifted, on-demand, over-the-top, and the like. The advertisements may be categorized as occurring in national or local/regional ad slots. The data from the smart TVs may be used to determine ad impressions, gross rating points, and target rating points. View rates for advertisements may also be determined and variations of advertisement may also be prepared.