Human Tracking via Body Joint Segmentation and Color Maps
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems for detecting and tracking humans in retail environments struggle to accurately distinguish customers from associates or employees in captured images, requiring manual intervention and lacking efficient automated solutions.
Innovation Solution
A system that uses a camera and control circuit to detect key body joints, segment body parts, and generate color distribution maps to differentiate between individuals, storing these maps for matching and tracking across images, allowing for automatic identification of retail associates and customers based on correlation values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual intervention is used to distinguish customers from associates in captured images, then identification accuracy can be maintained through human judgment, but labor intensity and processing time increase significantly
Solution Approach 1:
The system enables automatic self-service identification by having individuals register their body shape templates once, which are then automatically stored and used for future recognition. The control circuit automatically compares captured body shapes with stored templates without requiring manual intervention, allowing the system to serve itself in distinguishing customers from associates.
Solution Approach 2:
The patent replaces the mechanical human visual inspection process with an automated control circuit that captures images, extracts body shape features, and performs automatic comparison with stored templates. This substitution eliminates manual labor while maintaining identification accuracy through systematic automated processing.
2Productivity
If automated systems are implemented to track individuals across images, then processing efficiency increases, but system complexity and computational requirements worsen
Solution Approach 1:
The system segments the identification process into distinct components: body shape template creation, template storage, image capture, feature extraction, and comparison. By dividing the complex task of individual identification into these manageable segments, the system achieves high processing efficiency while keeping each component relatively simple and modular.
Solution Approach 2:
The system performs preliminary action by having individuals register their body shape templates in advance before actual tracking begins. These pre-stored templates serve as reference data that enable rapid automated comparison and identification during subsequent image processing, significantly improving processing efficiency without increasing real-time computational complexity.
3Measurement precision
If body shape templates are stored for each individual, then automatic recognition accuracy improves, but data storage requirements and processing complexity increase
Solution Approach 1:
The system extracts only the essential body shape template features from full images and stores only these extracted templates for later recognition. By taking out and storing only the critical identification features rather than complete images, the system achieves high recognition accuracy while minimizing data storage requirements and processing complexity.
Data Source
AI summary
In some embodiments, apparatuses and methods are provided herein useful to detecting and tracking humans. In some embodiments, there is provided a system for detecting and tracking humans from one image to another image including: a camera; a control circuit configure to: receive a first image; detect a plurality of key body joints of a first human captured on the first image; determine segmentations of the plurality of key body joints to determine one or more body parts of the first human; determine a color distribution map of aggregate pixels associated with each body part of the first human on the first image; and cause a database to store the color distribution map; and a database comprising one or more color distribution map sets each associated with a detected human in a captured image of the camera.