Picture Ranking via Social Relation Models for Face Images
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional picture ranking methods on terminals, such as smartphones, are monotonous as they primarily rely on time points and GPS locations, lacking diversity in grouping and ranking approaches.
Innovation Solution
A picture ranking method that detects whether acquired pictures contain human faces and ranks them using a social relation model if they do, or a preset rule if they do not, allowing for diverse ranking based on social connections or other criteria.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If pictures are ranked using conventional methods (time points and GPS locations), then the ranking is simple to implement, but the ranking manner is monotonous and lacks diversity
Solution Approach 1:
The patent segments the picture ranking process into different branches based on picture type: human face pictures are ranked using social relation models, while non-face pictures are ranked using conventional time and location methods. This segmentation allows the system to apply different ranking strategies to different picture types, thereby increasing ranking diversity without requiring complete system redesign.
Solution Approach 2:
The system integrates multiple ranking functions into a single unified framework. The terminal device can automatically select between social relation-based ranking and conventional time/location-based ranking based on the picture content, making the system multi-functional and adaptable to different picture types without requiring separate ranking systems.
2Adaptability or versatility
If social relation models are used to rank human face pictures, then picture grouping and ranking become more diverse and relevant, but the system complexity and computational requirements increase
Solution Approach 1:
The system automatically detects whether pictures contain human faces and autonomously selects the appropriate ranking method. For human face pictures, it automatically applies social relation models; for non-face pictures, it uses conventional methods. This self-service mechanism reduces the need for manual intervention and simplifies user interaction despite the underlying complexity.
Solution Approach 2:
The patent applies different quality levels of processing to different picture types. Human face pictures receive more sophisticated social relation-based processing, while non-face pictures use simpler conventional methods. This local quality approach ensures that computational resources are concentrated where they provide the most value (face recognition and social relation analysis) while maintaining efficiency for other picture types.
Data Source
AI summary
A picture ranking method and a terminal comprises acquiring pictures stored in a terminal, detecting whether the pictures are first-type pictures, where a first-type picture refers to a picture including a human face, and when the pictures are first-type pictures, ranking the pictures according to a social relation model, or when the pictures are not first-type pictures, ranking the pictures according to a preset rule.