Retail Display Control Using Clerk Recognition to Cut Processing Load
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing display technologies struggle to differentiate between customers and store clerks near a display device, leading to unnecessary attribute estimation and inappropriate information display for clerks, increasing computational burden and reducing efficiency.
Innovation Solution
A system that includes a camera for image acquisition, person extraction, determination of store clerk status based on facial images, and dynamic display control to differentiate between customers and clerks, using on-duty store clerk information for tailored display content.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If attribute estimation processing is performed for all persons photographed near the display device, then appropriate information can be displayed for customers, but the computational burden on the computer increases significantly
Solution Approach 1:
The system extracts and removes store clerks from the set of persons requiring attribute estimation processing. By identifying store clerks through facial recognition against on-duty store clerk information, the system excludes them from customer-facing attribute estimation, thereby reducing unnecessary computational load while maintaining appropriate information display for actual customers
Solution Approach 2:
The system applies different processing quality levels to different persons: store clerks receive identification and exclusion from processing, while customers receive full attribute estimation and tailored information display. This local differentiation optimizes computational resources by applying intensive processing only where needed
2Measurement precision
If attribute estimation processing is performed when store clerks are positioned near the display device, then the system can identify customers, but unnecessary processing increases the burden on the computer
Solution Approach 1:
The system performs preliminary identification of store clerks using facial recognition technology before proceeding with attribute estimation. By pre-identifying store clerks and excluding them from further processing, the system avoids unnecessary time-consuming attribute estimation operations while maintaining accurate person identification
3Adaptability or versatility
If the system processes all persons near the display device, then customer information can be displayed, but appropriate information cannot be displayed for store clerks
Solution Approach 1:
The system extracts store clerks from the general person processing pipeline and creates a separate handling path. By identifying store clerks through facial recognition and excluding them from customer-oriented attribute estimation, the system ensures that store clerk-specific information can be appropriately displayed without being contaminated by customer-focused processing
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The present invention provides a display control device (1) that acquires an image generated by a camera (4) installed near a display (3) installed in a store or in the display (3), extracts a person from the image, determines whether the person extracted from the image is a store clerk based on on-duty store clerk information including at least one of facial images and appearance feature amounts of store clerks (7) on duty, and changes display control of the display (3) according to whether the extracted person is a store clerk.