Designing a Contactless, AI System to Measure the Human Body using a Single Camera for the Clothing and Fashion Industry
Montazerian, Mohammad. 2024. Designing a Contactless, AI System to Measure the Human Body using a Single Camera for the Clothing and Fashion Industry. Doctoral thesis, Goldsmiths, University of London [Thesis]
|
Text (Designing a Contactless, AI System to Measure the Human Body using a Single Camera for the Clothing and Fashion Industry)
COM_thesis_MontazerianM_2024.pdf - Accepted Version Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (89MB) | Preview |
Abstract or Description
Using a single RGB camera to obtain accurate body dimensions rather than measuring these manually or via more complex multi-camera or more expensive 3D scanners, has a high application potential for the apparel industry.
In this thesis, a system that estimates upper human body measurements using a set of computer vision and machine learning techniques. The main steps involve: (1) using a portable camera; (2) improving image quality; (3) isolating the human body from the surrounding environment; (4) performing a calibration step; (5) extracting body features from the image; (6) indicating markers on the image; (7) producing refined final results.
In this research, a unique geometric shape is favored, namely the ellipse, to approximate human body main cross sections. We focus on the upper body horizontal slices (i.e. from head to hips) which, we show, can be well represented by varying an ellipse’s eccentricity, this per individual. Then, evaluating each fitted ellipse’s perimeter allows us to obtain better results than the current state-of-the-art for use in the fashion and online retail industry.
In our study, I selected a set of two equations, out of many other possible choices, to best estimate upper human body horizontal cross sections via perimeters of fitted ellipses. In this study, I experimented with the system on a diverse sample of 78 participants. The results for the upper human body measurements in comparison to the traditional manual method of tape measurements, when used as a reference, show ±1cm average differences, sufficient for many applications, including online retail.
Item Type: |
Thesis (Doctoral) |
Identification Number (DOI): |
|
Keywords: |
Human Body Measurement, Machine Learning, Computer Vision, Ellipse equations, Mobile Scanner, 3D Scanner, Fashion Technology |
Departments, Centres and Research Units: |
|
Date: |
29 February 2024 |
Item ID: |
35887 |
Date Deposited: |
08 Apr 2024 08:35 |
Last Modified: |
08 Apr 2024 12:01 |
URI: |
View statistics for this item...
Edit Record (login required) |