Automatic classification of female body shape using 3D anthropometric scan data

Area of Science:

• Anthropometry
• Apparel Design
• Data Science

Background:

• Conventional apparel sizing systems lack accuracy due to limited consideration of diverse female body shapes.
• Individual differences in body morphology are crucial for effective apparel fit and customization.
• Existing methods often fail to capture the full spectrum of human body diversity.

Purpose of the Study:

• To develop and validate a data-driven framework for classifying female body shapes using 3D body scan data.
• To establish a robust taxonomy of female body shapes based on anthropometric ratios.
• To provide a foundation for improved apparel pattern customization and mass-customized garment production.

Main Methods:

• Utilized 3D body scan data from 1,019 Korean women, extracting key anthropometric ratios (chest-to-hip, hip-to-waist, depth indices).
• Applied analysis of variance (ANOVA) and decision tree algorithms to derive body shape classification rules after data filtering.
• Employed canonical discriminant analysis for model validation, achieving high classification accuracy.

Main Results:

• Developed a body shape classification system identifying nine distinct types, consolidated into three macro groups: Top Hourglass, Regular Hourglass, and Bottom Hourglass.
• The classification model demonstrated high robustness with an overall accuracy of 93.6% using canonical discriminant analysis.
• The study established statistically grounded rules for reproducible body shape classification.

Conclusions:

• The proposed 3D scan-based methodology offers a statistically sound and reproducible approach to classifying female body shapes.
• This classification framework has significant implications for enhancing apparel fit and enabling mass-customized garment production.
• The findings contribute to a better understanding of morphological diversity for improved apparel industry practices.