Clothed Person Imaging Parameterization Using Body-Shape Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current imaging systems struggle to accurately determine body-shape information of clothed individuals, necessitating manual estimation and leading to suboptimal imaging configurations, especially in emergency scenarios where undressing is not feasible.
Innovation Solution
A computer-implemented method using a trained machine-learning model (MLM) analyzes image data of a clothed person to predict body-shape information, which is then used to automatically determine imaging parameters, thereby enhancing the accuracy and automation of imaging systems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual palpation is used to estimate body-shape information, then the imaging system can be configured, but the accuracy of body-shape estimation deteriorates and operational efficiency decreases
Solution Approach 1:
The patent replaces the mechanical manual palpation process with an automated optical imaging and machine learning system. Cameras capture images of the clothed person, and a trained MLM automatically predicts body-shape information, eliminating the need for manual mechanical estimation while improving both accuracy and operational efficiency.
Solution Approach 2:
The system enables self-service by allowing the imaging system to automatically determine body-shape information without requiring manual intervention. The trained MLM processes images and outputs body-shape predictions autonomously, making the system self-sufficient in acquiring necessary measurement data.
2Extent of automation
If camera images are used for rough pre-initialization, then automation is improved, but measurement precision deteriorates due to clothing interference
Solution Approach 1:
The patent changes the approach from direct visual measurement to indirect prediction through machine learning. Instead of attempting to directly measure body shape from clothed images (which is imprecise), the system uses the trained MLM to learn the mapping between clothed appearance and underlying body shape, transforming the measurement problem into a predictive modeling task.
Solution Approach 2:
The trained machine learning model acts as an intermediary between the clothed person's visual appearance and the actual body-shape information. The MLM processes the image data and translates it into accurate body-shape predictions, serving as a mediator that overcomes the obstacle of clothing interference.
3Ease of operation
If imaging parameters are manually adjusted, then the imaging system can be configured, but the complexity of operation increases and time consumption increases
Solution Approach 1:
The system performs preliminary action by pre-training the machine learning model on extensive datasets before actual use. This preliminary training enables the system to automatically make accurate predictions during operation, eliminating the need for manual parameter adjustment and reducing configuration time while maintaining ease of operation.
Data Source
AI summary
In a computer-implemented method for parameterizing an imaging system for mapping a clothed person, image data about the clothed person is obtained and body-shape information about the person is determined by applying a trained machine-learning model to the image data. At least one imaging parameter of the imaging system is determined as a function of the body-shape information.