Classification and recognition of young males’ waist-abdomen-hip shape based on body photos

• Published in: Industria textilă (Bucharest, Romania : 1994) Vol. 73; no. 5; pp. 511 - 518
• Main Authors: XU, KAI-YI, ZHANG, JIAN, ZHAO, SONG-LING, WANG, RUO-WEN, GU, BING-FEI
• Format: Journal Article
• Language: English
• Published: Bucharest The National Research & Development Institute for Textiles and Leather - INCDTP 01.01.2022
• Subjects: Abdomen; Body measurements; Classification; Clothing industry; Customization; Human body; Parameters; Position measurement; Shape recognition; Three dimensional models
• ISSN: 1222-5347
• DOI: 10.35530/IT.073.05.202160

To improve somatotype and realize rapid customization of clothing, this study proposed a body-shape recognition method based on the front and side body photos. The shape parameters such as angles, heights, lengths, and ratios at the waist-abdomen-hip position were measured or calculated based on the manual and three-dimensional (3D) point cloud data of 180 young men. Through analysis, four morphological parameters were determined to classify the waist-abdomen-hip shape and establish the classification rules of the four body types (fat, normal, forward fat and obese). Finally, according to the front and side photos of the human body, the orthogonal silhouettes were extracted to obtain the four parameters that can be used to distinguish the waist-abdomen-hip body shape of young men, and the photo-based recognition method of the waist-abdomen-hip body shape could be realized automatically. The verification results showed that the recognition accuracy ratio reaches 93.3%, indicating that the waist-abdomen-hip shape identification system using this body-shape recognition method based on body photos is effective and can provide a basis for personalized clothing customization to satisfy the individual needs of the consumers.