Preference Service System Using Face and Posture Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for providing preference-based services struggle to accurately analyze user preferences due to the lack of consideration for emotional and attention states, and the limited accuracy in tracking user interests without comprehensive data, such as posture information.
Innovation Solution
A system and method that utilize face and posture image analysis through deep neural networks to extract user characteristic information, including emotional states, attention, gender, and age, to provide personalized services by setting priorities based on these factors, incorporating history and bio-signal data for improved accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If only watching history and frequency are used for preference analysis, then the system is simple to operate, but the accuracy of preference extraction is low
Solution Approach 1:
The patent combines multiple data sources including watching history, face images, posture images, and bio-signal data into a unified preference analysis system. This integration allows the system to extract more accurate user preferences by synthesizing information from diverse channels rather than relying on a single data source.
Solution Approach 2:
The patent introduces deep neural networks as intermediary components that process and analyze raw data from cameras and sensors. These neural networks serve as mediators that transform complex raw data into meaningful preference indicators, enabling accurate analysis without requiring the system to directly handle all raw data complexity.
2Measurement precision
If only line of sight information from front camera is used, then the device complexity is low, but the accuracy of user interest analysis is insufficient
Solution Approach 1:
The patent merges data from multiple sensing modalities including front camera line of sight, rear camera face images, posture detection, and bio-signal sensors. This multi-source data fusion comprehensively captures user interest by combining visual attention data with physiological responses and body language, significantly improving analysis accuracy.
Solution Approach 2:
The patent adds new dimensions to user analysis by incorporating spatial information from rear cameras and temporal information from bio-signal data. This multi-dimensional approach moves beyond simple line of sight tracking to include facial expressions, posture dynamics, and physiological states, creating a more complete picture of user interest.
3Reliability
If comprehensive data including face and posture information is collected, then preference analysis accuracy is improved, but the device complexity increases
Solution Approach 1:
The patent segments the complex preference analysis system into distinct functional modules: face image acquisition module, posture detection module, bio-signal collection module, and preference calculation module. Each module handles specific data types and processing tasks independently, making the overall complex system more manageable and maintainable through modular architecture.
Solution Approach 2:
The patent employs deep neural networks as intermediary processing layers that receive raw data from multiple complex sources and output simplified preference metrics. These neural network intermediaries absorb the complexity of data integration and processing, presenting a simplified interface to the rest of the system while handling the computational burden of multi-source data fusion.
Data Source
AI summary
Disclosed is a preference-based service providing method for operating preference-based service providing system and device by executing an artificial intelligence (AI) algorithm and/or a machine learning algorithm in a 5G environment connected for the Internet of things. A preference-based service providing method according to an embodiment of the present disclosure may include acquiring user video information obtained by imaging a user who is using an electronic device, analyzing a preference of the user for a service provided by the electronic device on the basis of the user video information including a face image and a posture image of the user, setting a priority of the service provided by the electronic device on the basis of the preference of the user, and providing a recommendation list of services provided by the electronic device on the basis of priorities of the services.