Cross-Camera Object Tracking Using Biometric Identifiers
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional tracking systems require cumbersome setup processes and high processing and memory requirements, especially with multiple cameras, making them inefficient for tracking objects across different video streams.
Innovation Solution
A video surveillance system that generates a unique biometric identifier for an object of interest, allowing seamless tracking across multiple video streams by comparing candidate identifiers and automatically switching the primary viewing stream to the object's current location.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional tracking systems are used with multiple cameras, then object tracking capability is provided, but setup complexity and processing requirements increase significantly
Solution Approach 1:
The system segments the tracking task by assigning each camera its own video stream processing, extracting features independently at each camera location. This distributed approach reduces the complexity of coordinating multiple cameras while maintaining comprehensive object tracking capability across the entire environment.
Solution Approach 2:
The system introduces a feature database as an intermediary that stores extracted features from each camera. This database acts as a mediator between cameras, enabling object tracking across multiple views without requiring direct coordination between cameras or complex pre-programming of movement vectors.
2Reliability
If conventional tracking systems process multiple video streams, then comprehensive object monitoring is achieved, but processing power and memory requirements increase
Solution Approach 1:
The system extracts only the essential features from each video stream (such as object appearance, position, and characteristics) rather than processing the entire video data. These extracted features are stored in a database and compared to identify tracked objects, significantly reducing processing power and memory requirements while maintaining reliable comprehensive monitoring.
Solution Approach 2:
The system performs partial processing by extracting and storing only the necessary feature information from each camera stream rather than processing all video data in detail. This partial action approach reduces computational burden while sufficient for reliable object tracking across multiple cameras.
3Area of stationary object
If cameras are positioned to cover all areas with minimal overlap, then comprehensive coverage is achieved, but system setup becomes cumbersome
Solution Approach 1:
The system makes each camera multi-functional by enabling any camera to potentially capture the object of interest. Rather than requiring precise coordination and specific positioning, the system allows cameras to be placed more flexibly while still achieving comprehensive coverage through the feature extraction and database comparison mechanism that works across all camera views.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Example implementations include a method, apparatus and computer-readable medium in a video surveillance system for tracking an of-interest (OI) object, comprising receiving, from a user interface, a request to track an object in a first video stream from a plurality of video streams captured from a first camera of a plurality of cameras installed in an environment. The implementations further include extracting at least one image of the object from the first video stream. Additionally, the implementations further include generating a unique biometric identifier of the object based on the at least one image. Additionally, the implementations further include detecting, using the unique biometric identifier, the object in a second video stream captured from a second camera of the plurality of cameras, and outputting, on the user interface, the second video stream in response to detecting the object in the second video stream.