Search citation statements
Paper Sections
Citation Types
Year Published
Publication Types
Relationship
Authors
Journals
PurposeTo enhance the fit, comfort and overall satisfaction of lower body attire for online shoppers, this study introduces a reclassification method of the lower body profiles of young females in complex environments, which is used in the framework of remote clothing mass customization.Design/methodology/approachFrontal and lateral photographs were collected from 170 females prior, marked as size M. Employing a salient object detection algorithm suitable for complex backgrounds, precise segmentation of body profiles was achieved while refining the performance through transfer learning techniques. Subsequently, a skeletal detection algorithm was employed to delineate distinct human regions, from which 21 pivotal dimensional metrics were derived. These metrics underwent clustering procedures, thus establishing a systematic framework for categorizing the lower body shapes of young females. Building upon this foundation, a methodology for the body type combination across different body parts was proposed. This approach incorporated a frequency-based filtering mechanism to regulate the enumeration of body type combinations. The automated identification of body types was executed through a support vector machine (SVM) model, achieving an average accuracy exceeding 95% for each defined type.FindingsYoung females prior to being marked as the same lower garment size can be further subdivided based on their lower body types. Participants' torso types were classified into barrel-shaped, hip-convex and fat-accumulation types. Leg profile shapes were categorized into slender-elongated and short-stocky types. The frontal straightness of participants’ legs was classified as X-shaped, I-shaped and O-shaped types, while the leg side straightness was categorized based on the knee hyperextended degree. The number of combinations can be controlled based on the frequency of occurrence of combinations of different body types.Originality/valueThis methodological advancement serves as a robust cornerstone for optimizing clothing sizing and enabling remote clothing mass customization in E-commerce, providing assistance for body type database and clothing size database management as well as strategies for establishing a comprehensive remote customization supply chain and on-demand production model.
PurposeTo enhance the fit, comfort and overall satisfaction of lower body attire for online shoppers, this study introduces a reclassification method of the lower body profiles of young females in complex environments, which is used in the framework of remote clothing mass customization.Design/methodology/approachFrontal and lateral photographs were collected from 170 females prior, marked as size M. Employing a salient object detection algorithm suitable for complex backgrounds, precise segmentation of body profiles was achieved while refining the performance through transfer learning techniques. Subsequently, a skeletal detection algorithm was employed to delineate distinct human regions, from which 21 pivotal dimensional metrics were derived. These metrics underwent clustering procedures, thus establishing a systematic framework for categorizing the lower body shapes of young females. Building upon this foundation, a methodology for the body type combination across different body parts was proposed. This approach incorporated a frequency-based filtering mechanism to regulate the enumeration of body type combinations. The automated identification of body types was executed through a support vector machine (SVM) model, achieving an average accuracy exceeding 95% for each defined type.FindingsYoung females prior to being marked as the same lower garment size can be further subdivided based on their lower body types. Participants' torso types were classified into barrel-shaped, hip-convex and fat-accumulation types. Leg profile shapes were categorized into slender-elongated and short-stocky types. The frontal straightness of participants’ legs was classified as X-shaped, I-shaped and O-shaped types, while the leg side straightness was categorized based on the knee hyperextended degree. The number of combinations can be controlled based on the frequency of occurrence of combinations of different body types.Originality/valueThis methodological advancement serves as a robust cornerstone for optimizing clothing sizing and enabling remote clothing mass customization in E-commerce, providing assistance for body type database and clothing size database management as well as strategies for establishing a comprehensive remote customization supply chain and on-demand production model.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.