Skeleton-Based Customer Interest Detection From Retail Behavior Images
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies require manual generation of detection rules for each commodity product to effectively detect customer interest, which is impractical due to the large number of products, making it difficult to provide efficient customer service.
Innovation Solution
An information processing system that generates detection rules based on past behavior and degree of interest, associating customer behaviors with commodity product features without relying on individual product-specific rules, using cameras and POS data to track and analyze customer interactions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual detection rules are generated for each commodity product, then detection accuracy for customer interest is improved, but the time and effort required increases significantly
Solution Approach 1:
The patent segments the detection rule generation process into automated components that analyze past customer behavior data and generate rules systematically, rather than requiring manual creation for each product. This segmentation allows the system to handle multiple products simultaneously through automated data processing.
Solution Approach 2:
The system enables self-service by automatically generating detection rules using past behavior data without requiring manual intervention for each commodity product. The automated rule generation process serves itself by utilizing existing data infrastructure to create detection criteria across all products.
2Adaptability or versatility
If detection rules are customized for each commodity product, then customer service effectiveness is improved, but the complexity of rule generation increases
Solution Approach 1:
The patent applies universality by creating a single automated detection rule generation system that serves all commodity products simultaneously. The system uses a unified approach to analyze behavior data and generate applicable rules across diverse product categories, eliminating the need for separate customization processes.
Solution Approach 2:
The system changes parameters by utilizing past behavior data as the basis for rule generation, transforming the approach from manual specification to data-driven automatic generation. This parameter change enables the system to adapt to different products by analyzing their specific behavior patterns without increasing operational complexity.
3Measurement precision
If comprehensive behavior analysis is performed for each product, then detection precision is improved, but the productivity of the detection system decreases
Solution Approach 1:
The patent ensures continuity of useful action by implementing an automated system that continuously generates and updates detection rules based on accumulating past behavior data. This continuous operation maintains high detection precision across all products simultaneously, improving overall system productivity compared to manual methods.
Solution Approach 2:
The system replaces the mechanical process of manual rule creation with an automated computational system that generates detection rules algorithmically. This substitution maintains comprehensive behavior analysis for precision while dramatically improving productivity by eliminating manual intervention across all product lines.
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
An information processing apparatus (50) detects (62) a person and a commodity product from image data. The information processing apparatus (50) acquires (62) skeleton information on the detected person. The information processing apparatus (50) specifies (63), based on the skeleton information on the person, a behavior of the person exhibiting with respect to the commodity product. The information processing apparatus (50) specifies (64), based on the specified behavior of the person exhibiting with respect to the commodity product, a combination of an attribute of the commodity product and a degree of interest in the commodity product.