Vehicle Body Parameter Estimation Using Dynamic Posture Capture
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining body parameters of vehicle occupants are limited, as they often rely on a single posture capture and can be obscured by clothing, requiring additional equipment like IR cameras, which are costly and inconvenient.
Innovation Solution
A method using a single omnidirectional camera integrated into the vehicle, capturing data from multiple postures to estimate body parameters like height, arm length, and torso length, without the need for additional equipment, by employing anthropometric models and iterative data capture to account for clothing obscuration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single data capturing device is used to capture body parameters, then device complexity is reduced, but measurement precision deteriorates due to obscured body parts
Solution Approach 1:
The system captures images at multiple time points as the person moves through different body postures (approaching, standing still, moving away). This dynamic capture approach allows the same camera to observe different body parts at different times, resolving the contradiction between using a single device and achieving complete body measurement.
Solution Approach 2:
The system adds the time dimension to the capture process by taking multiple images at different time points. This transforms a static single-view problem into a dynamic multi-view problem, enabling complete body parameter extraction from a single spatial location through temporal variation.
2Measurement precision
If multiple data capturing devices are used to capture all body parts, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The single data capturing device performs multiple functions: capturing the person at different distances, different body postures, and different angles by utilizing the person's natural movements. This eliminates the need for multiple dedicated cameras for different body parts.
Solution Approach 2:
The system utilizes the person's own movements and posture changes as the source of varied capture angles and views. The person's natural behavior during approach and departure provides the diversity of views needed, without requiring additional active components or multiple devices.
3Ease of operation
If anatomical data is captured for only one body posture, then ease of operation is improved, but measurement precision deteriorates due to obscured body parts
Solution Approach 1:
Instead of requiring the person to maintain a fixed posture, the system captures images dynamically as the person moves through different postures naturally. This maintains operational simplicity while ensuring all body parts become visible at some point during the sequence.
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
The present disclosure relates to a method for determining a body parameter (A, L, BH, EP, T) of a person (8), the person (8) being outside a vehicle (2). The method comprises the steps of: a) capturing a first set of data of the person (8) by means of a first data capturing device (6) of the vehicle (2), the captured first set of data being representative of a first body posture of the person (8), b) capturing a second set of data of the person (8) by means of a second data capturing device (6) of the vehicle (2), the captured second set of data being representative of a second body posture of the person (8), the second body posture being different from the first body posture, c) using the first and second sets of data as input for estimation of the body parameter (A, L, BH, EP, T) of the person (8). The present disclosure also relates to the use of a data capturing device of a vehicle, and optionally a distance measurement system of the vehicle, for determining a body parameter of a person according to the method.