Biometric Emotion Recognition for Intelligent Artwork Recommendation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing electronic frames lack intelligence in artwork recommendation, relying mainly on commercial value and viewing history, which is not user-friendly and fails to consider the user's emotional state.
Innovation Solution
An emotion recognition-based artwork recommendation method using deep learning algorithms to determine a user's current emotion by analyzing biometric parameters such as facial and sound features, selecting and displaying artwork corresponding to the user's emotional state, and recommending images based on emotional duration and historical data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If traditional artwork recommendation methods based on commercial value and viewing history are used, then the recommendation system is simple to implement, but the intelligence degree and user-friendliness are low
Solution Approach 1:
The system performs preliminary classification training on multiple biometric parameters using deep learning algorithms to build emotion classification models in advance. This preparation work enables the system to automatically recognize user emotions and provide intelligent artwork recommendations without complex real-time processing during actual use.
Solution Approach 2:
The patent introduces biometric parameters (facial features, sound features, etc.) as intermediary elements between the user and the artwork recommendation system. These parameters serve as mediators that convey user emotional states to the system, enabling intelligent recommendations without requiring direct complex user-system interaction.
2Adaptability or versatility
If emotion recognition based on multiple biometric parameters is implemented, then the user experience and personalization are improved, but the processing complexity and computational requirements increase
Solution Approach 1:
The patent segments the emotion recognition process into multiple independent classification tasks, each handling a specific biometric parameter (facial features, sound features, etc.). Multiple emotion classification sub-models are trained separately for different parameter classifications, and their results are integrated to determine the user's current emotion type. This segmentation reduces the complexity of each individual model while maintaining comprehensive emotion recognition capability.
Solution Approach 2:
The system transforms complex biometric parameter data into simplified emotion type classifications through deep learning processing. By converting raw biometric data into discrete emotion categories, the system reduces processing complexity while preserving the essential emotional information needed for personalized artwork recommendations.
3Measurement precision
If real-time biometric parameter analysis is performed to determine current emotion type, then the recommendation accuracy is improved, but the processing time and computational resources increase
Solution Approach 1:
The system performs classification training processing on biometric parameters in advance to build pre-trained emotion classification models. During real-time operation, these pre-trained models can quickly and accurately determine user emotions without requiring extensive computational resources, thus reducing processing time while maintaining high recognition accuracy.
Data Source
AI summary
The present disclosure provides an emotion recognition-based artwork recommendation method and device. The method includes: obtaining a current biometric parameter of a user; determining a current emotion type of the user according to the current biometric parameter; selecting an image of an artwork corresponding to the current emotion type according to the current emotion type; and recommending an image of the artwork to the user by displaying the image of the artwork on the display screen.