Multi-Camera Person Tracking Interface for Faster Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Tracking a specific person across multiple videos captured by multiple cameras requires significant human effort and time, as users need to manually locate and identify the person in each video.
Innovation Solution
A person tracking support device that includes a storage device and processing circuitry to identify persons in videos, display a synchronized multi-camera video playback screen with bounding boxes, and provide a second display section for tracking information, allowing users to efficiently designate and recognize the target person's appearance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual tracking of a person across multiple videos is performed, then tracking accuracy can be maintained, but the time cost and human effort increase significantly
Solution Approach 1:
The system performs automatic person identification and tracking across multiple videos using AI technology, allowing the system to serve itself rather than requiring manual intervention. The processing circuitry automatically detects persons, identifies them across different videos, and displays tracking information without human effort, thereby resolving the contradiction between tracking accuracy and time cost.
Solution Approach 2:
The patent replaces manual mechanical tracking operations with automated AI-based person detection and recognition systems. The processing circuitry uses machine learning models to automatically identify persons across videos, substituting human manual labor with automated computational processes, thus reducing time cost while maintaining tracking accuracy.
2Reliability
If manual location and identification of persons in each video is performed, then tracking reliability is maintained, but productivity decreases due to high human cost
Solution Approach 1:
The system automatically performs person detection, identification, and tracking across multiple videos without requiring human intervention. The processing circuitry self-services by executing AI-based algorithms to maintain tracking reliability while significantly improving productivity, as the system works continuously without human labor constraints.
Solution Approach 2:
The patent changes the operational parameters from manual human processing to automated computational processing. By transitioning from human-based tracking to AI-based tracking, the system maintains reliability through consistent algorithmic application while dramatically improving productivity through automated high-speed processing.
3Ease of operation
If synchronized multi-camera video playback is displayed with tracking information, then ease of operation is improved, but device complexity increases
Solution Approach 1:
The display interface is segmented into multiple functional sections: a video playback section for displaying synchronized multi-camera footage and a separate tracking information section for displaying person identification details. This segmentation organizes complex information into manageable visual components, improving ease of operation while the underlying processing complexity remains managed through modular system architecture.
Data Source
AI summary
A person tracking support device identifies persons appearing in at least one of videos, and displays a person tracking screen including a first display section that synchronously reproduces monitoring videos selected from the videos and a second display section that displays an appearance section in which the person appears in at least one of the monitoring videos for each persons. When a user performs an operation of designating a time point of the appearance section related to a target person on the person tracking screen, the person tracking support device displays a snapshot of the monitoring videos at the time point on the first display section, and displays person tracking information for the user to recognize a portion in which the target person appears in the snapshot on the person tracking screen.