Onlooker Detection Using Distance, Face, and Key Point Cues
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing onlooker detection systems using depth maps often generate false alarms when a person passes by without peeping behavior, leading to inaccurate security classifications.
Innovation Solution
An onlooker detection system and method that utilizes a person detection module to obtain distance information and an onlooker determination module to classify individuals based on their proximity and orientation relative to a device, incorporating face and key point information to determine security classifications such as onlooker, passerby, or sharing user categories.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If depth map detection is used to identify onlookers, then security monitoring capability is improved, but false alarm rate increases due to inability to distinguish passing persons from actual onlookers
Solution Approach 1:
The detection system segments the analysis into multiple independent components: depth information detection, face detection, and key point detection. Each component processes specific features separately, allowing comprehensive evaluation of person behavior through combined results, thereby reducing false alarms while maintaining security monitoring capability
Solution Approach 2:
The system transitions from single-dimensional depth map detection to multi-dimensional detection by incorporating face detection and key point detection dimensions. This dimensional expansion enables more accurate distinction between passing persons and actual onlookers through comprehensive feature analysis
2Device complexity
If only depth information is used for onlooker detection, then system complexity is reduced, but detection accuracy deteriorates due to false alarms
Solution Approach 1:
The detection process is segmented into distinct modules: depth information processing, face detection, and key point detection. Each module handles specific tasks independently, making the complex multi-feature detection system manageable while achieving high detection accuracy through integrated results
Solution Approach 2:
The detection system achieves multi-functionality by using a unified detection framework that processes depth information, face features, and key point features simultaneously. This universal approach enables accurate distinction between different person types without requiring separate specialized systems
Data Source
AI summary
An onlooker detection system and an onlooker detection method are provided. The onlooker detection system includes: a person detection module, configured to receive an image, and obtain, in response to presence of persons in the image, person information of each person, where the person information includes distance information relative to a device; and an onlooker determination module, configured to: determine whether the persons include at least one non-user present in a range based on the distance information of the person information of each person; and determine, in response to presence of the at least one non-user in the range, a security classification to which each non-user belongs based on the person information of each non-user, where the security classification includes an onlooker category.