Egocentric Item Detection for Camera-Light Unmanned Store Checkout
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing unmanned store payment systems require numerous cameras, leading to high installation and operating costs, and lack accessibility for visually impaired individuals, due to limited camera coverage and complex computation requirements.
Innovation Solution
An in-store automatic payment method using a mobile terminal to collect egocentric video, detect purchase items, calculate reliability, and provide device focus navigation, reducing the need for fixed cameras and enhancing accessibility for visually impaired users through egocentric video processing and reinforcement learning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If numerous fixed cameras are installed to achieve accurate item detection and automatic payment, then measurement precision is improved, but device complexity and installation expense increase
Solution Approach 1:
The mobile terminal performs self-service by capturing egocentric videos and detecting items autonomously without requiring external fixed cameras. The device uses its own computing resources and sensors to identify purchase target items, eliminating the need for numerous store-installed cameras while maintaining detection accuracy
Solution Approach 2:
Instead of having fixed cameras monitor customers from above, the system inverts the approach by having customers actively capture videos themselves using their mobile terminals. This role reversal transforms the detection system from passive surveillance to active first-person documentation, reducing infrastructure requirements
2Measurement precision
If numerous fixed cameras are installed to cover the entire store, then measurement precision is improved, but installation expense and operating expense increase
Solution Approach 1:
Customers use their personal mobile terminals for item detection, transferring the detection function from store infrastructure to customer devices. This eliminates the need for expensive camera installations and reduces operating expenses related to camera maintenance and data processing
Solution Approach 2:
Instead of installing expensive camera systems in every store, the patent uses copies of detection algorithms running on customers' mobile terminals. Each terminal becomes a portable detection unit, eliminating the need for duplicate camera installations across multiple locations
3Device complexity
If egocentric video processing and device focus navigation are implemented to reduce fixed cameras, then device complexity is reduced, but measurement precision may deteriorate
Solution Approach 1:
The system dynamically adjusts the mobile terminal's camera focus and positioning through device focus navigation, which guides users to optimal viewing angles and distances. This dynamic adjustment ensures high-quality item capture despite the absence of fixed cameras, maintaining detection precision while reducing system complexity
Solution Approach 2:
The system implements feedback mechanisms where detection reliability is continuously assessed and used to guide further navigation or request additional video captures. This feedback loop ensures that item detection accuracy is maintained by adaptively adjusting the collection process based on real-time reliability assessments
Data Source
AI summary
The present disclosure relates to an in-store automatic payment method and system applied to an unmanned store service, wherein the method and the system reduce the burden of excessively collecting videos from a number of fixed cameras installed in the related art by using an egocentric video, computer computation overhead, and installation expense and operating expense of a system for an unmanned store. The method includes: collecting, by a mobile terminal, an egocentric video; detecting, by at least one among multiple devices, a purchase target item from the egocentric video; calculating, by at least one among the multiple devices, a level of reliability for the purchase target item from the egocentric video; and registering, by at least one among the multiple devices, the purchase target item on a selected-item list of a user when the level of reliability is equal to or greater than a preset threshold.