Video Ad Effect Evaluation via Facial Micro-Movement Similarity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for evaluating advertising effects on video content are costly, prone to response bias, and subjective, with gaze-tracking methods being ambiguous and inaccurate due to the need for separate equipment and conscious viewer responses.
Innovation Solution
A method and system that evaluates advertising effects by recognizing unconscious viewer reactions through facial micro-movements, calculating similarity between character and viewer micro-movements, and determining an advertising effect score using facial micro-movement data extracted from specific regions of interest, filtered, and transformed into time-series power spectral densities for cross-entropy analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If survey methods are used to evaluate advertising effects, then viewer feedback can be obtained, but high resource costs and response bias occur
Solution Approach 1:
The patent replaces the mechanical survey system (questionnaires, interviews) with an automated computer vision system that captures facial micro-movements. This substitution eliminates the need for human respondents and manual data collection, thereby reducing resource investment while maintaining or improving measurement accuracy through objective physiological data.
Solution Approach 2:
The system enables self-service measurement by automatically capturing and analyzing viewer facial expressions without requiring active participation from viewers. The evaluation process happens passively as viewers naturally watch content, eliminating the need for organized surveys and reducing resource requirements for data collection.
2Measurement precision
If gaze tracking equipment is used, then viewer attention can be monitored, but measurement accuracy decreases due to subjective judgment
Solution Approach 1:
The patent extracts the measurement function from complex specialized gaze-tracking equipment and integrates it into standard video playback systems. By using conventional cameras and computer vision algorithms to detect facial micro-movements, the system removes the need for separate tracking devices while achieving more objective and accurate measurements of viewer engagement.
Solution Approach 2:
Instead of using expensive specialized gaze-tracking hardware, the system creates a functional copy using standard video cameras and image processing techniques. The computer vision algorithms replicate the attention-monitoring capability of specialized equipment while using readily available components, thereby reducing device complexity.
3Loss of information
If facial micro-movement analysis is performed, then unconscious viewer reactions can be detected, but data processing complexity increases
Solution Approach 1:
The patent segments the facial analysis process into distinct components: face detection, landmark identification, micro-movement extraction, and emotional state classification. This segmentation allows each component to be optimized independently and processed through modular algorithms, reducing overall system complexity while maintaining the ability to detect unconscious reactions.
Solution Approach 2:
The system focuses on analyzing only the most informative facial regions (key landmarks such as eyes, eyebrows, and mouth) rather than processing entire facial images. By concentrating computational resources on critical micro-movement zones, the system achieves accurate unconscious reaction detection with reduced processing complexity.
Data Source
AI summary
The disclosure is related to a method and system for evaluating advertising effects of video content. The evaluation method includes presenting video content including a character to a viewer through a display, extracting pieces of facial micro-movement data (MMD) of the character in the video content and the viewer, while the viewer watches the video content, calculating a similarity of the MMD of the character and the MMD of the viewer, and calculating an advertising effect score of the video content on the basis of the similarity.