Garment fit evaluation using neural networks technology

Abstract

At present, garment e-commerce is developing rapidly, and the number of online shopping has increased significantly. In this context, the intelligent evaluation technology of garment fit is particularly important. Fit is the most basic requirement in the process of human dressing. It is of great significance to study the relationship between garment fit and human movement, garment structure, and fabric properties. In this article, we propose a garment fit evaluation model based on back propagation artificial neural network. This method realizes the evaluation of garment fit without any tryout. The inputs of the model are the anthropometric data, garment pattern and fabric properties, while the output is the prediction result of garment fit (fit or unfit). In order to build and train the model, the input and output data were obtained by experiment. And a total of 284 experimental samples were obtained. Through the real try-on test, the results revealed that this approach can effectively evaluate the fit of garment. It introduces new ideas and methods for the intelligent evaluation of garment fit, and has a certain reference value for the research of intelligent evaluation technology of garment fit.

Keywords:

• Fit evaluation
• data learning
• artificial neural networks

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

The work was financially supported by The National Natural Science Foundation of China, China (No. 61806161), Innovation ability support plan of Shaanxi Province-young science and Technology Star Project, China (No. 2020KJXX-083), Science and technology guidance project of China Textile Industry Federation, China (No. 2019049), China and The Youth Innovation Team of Shaanxi Universities, China.