Cornell University
Library
Cornell UniversityLibrary

eCommons

Help
Log In(current)
  1. Home
  2. Cornell University Graduate School
  3. Cornell Theses and Dissertations
  4. Dress Style Recommendation Based on Female Body Shapes

Dress Style Recommendation Based on Female Body Shapes

File(s)
Zong_cornell_0058O_11428.pdf (2.56 MB)
Permanent Link(s)
https://doi.org/10.7298/w7df-en12
https://hdl.handle.net/1813/112171
Collections
Cornell Theses and Dissertations
Author
Zong, Wenjia
Abstract

Fashion is an important part of life. According to Kant, fashion and style are the fundamental expressions of aesthetic tastes in the societal format. Clothing styles and apparel fit are the two key factors when consumers make purchase decisions. People have different body shapes and fit preferences; likewise, people have different aesthetic tastes in clothing design and styles. Many designer clothes are based on standard size systems, i.e., measurements, grading, pattern making, or common hourglass dress form. However, body shape is a complex physical attribute that can be gauged to help consumers achieve better aesthetic fitting styles for their needs. Therefore, this research aimed to develop a body shape-based style recommendation system, which could provide desirable apparel silhouettes and styles to consumers based on the relationships between dress attributes and body shape attributes. The recommended dress styles were collected from online channels and interviews with professional stylists. The Female Figure Identification Technique (FFIT) was adopted for the body shape categorization and validated with body measurement from SizeUSA. A style-by-body shape recommendation system was implemented on an interactive website with 3D dress rendered dresses and body shapes. The proposed recommendation system was examined by 171 fashion consumers to validate whether the recommendation would satisfy their need of personal style, body shape, and have an impact on their purchase intentions. A dress style recommendation system was developed based on surveys from online media and stylists’ interviews. Although no significance was found between the experimental (recommended) and control (all-option) conditions, the findings indicated that different body shapes had distinctive dress style preferences on waistlines and silhouettes.

Description
131 pages
Date Issued
2022-08
Keywords
3D virtual fashion
•
Body shape
•
Data science
•
Fashion algorithm
•
Style recommendation system
Committee Chair
Baytar, Fatma
Committee Member
Belongie, Serge J.
Degree Discipline
Fiber Science and Apparel Design
Degree Name
M.A., Fiber Science and Apparel Design
Degree Level
Master of Arts
Type
dissertation or thesis
Link(s) to Catalog Record
https://newcatalog.library.cornell.edu/catalog/15578989

Site Statistics | Help

About eCommons | Policies | Terms of use | Contact Us

copyright © 2002-2026 Cornell University Library | Privacy | Web Accessibility Assistance