A novel evaluation technique for human body perception of clothing fit - Université de Lille Accéder directement au contenu
Article Dans Une Revue Multimedia Tools and Applications Année : 2023

A novel evaluation technique for human body perception of clothing fit

Résumé

Fit evaluation plays an important role in garment products development and sales. Effective clothing fit evaluation methods can reduce the development cost of apparel products and the return rate of online apparel sales. In this research, we proposed an intelligent fit evaluation technology to predict clothing fit. The mathematical relationship model between clothing fit levels and indexes reflecting the clothing fit levels was constructed by using decision tree C4.5 algorithm. Then, two experiments were carried out to collect input and output training data. After learning from the collected data, the proposed model can predict clothing fit accurately. Next, we validated our proposed model’s prediction accuracy using K-fold cross validation. Finally, we gave two applications of the proposed model for clothing products development and shopping online. Results show that our proposed method has high prediction accuracy and less requirement for the number of learning samples, and can predict clothing fit automatically and rapidly without real try-on.
Fichier non déposé

Dates et versions

hal-04467614 , version 1 (20-02-2024)

Identifiants

Citer

Kaixuan Liu, C. Zhu, Xuyuan Tao, Pascal Bruniaux, Xianyi Zeng, et al.. A novel evaluation technique for human body perception of clothing fit. Multimedia Tools and Applications, 2023, Multimedia Tools and Applications, 82, pp.21057-21069. ⟨10.1007/s11042-023-14530-x⟩. ⟨hal-04467614⟩

Collections

UNIV-LILLE
10 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More