服装外轮廓识别与量化的研究进展

Research Progress in Garment Contours Recognition and Quantification

服装的外轮廓量化是服装智能化设计与生产的前提条件。介绍了服装外轮廓量化的过程,以及传统量化方法与基于人工智能的量化方法。概述了基于特征参数识别和数学公式计算的服装外轮廓量化方法,探究基于扫描等技术构建三维服装模型的外轮廓量化研究方法,并对其优缺点进行分析;总结基于机器学习与生成网络等方法的服装外轮廓量化。服装外观廓形具有复杂柔性多变的特征,目前对服装外轮廓量化的研究大多以简单的服装为样本,应用在廓形分类等领域;在实现服装各部位具体数值推理判断方面的应用相对不足,多为定性研究。研究认为,未来可以综合考虑外轮廓造型与服装款式及面料等多方面因素,并与人工智能技术相结合,使其向智能化、自动化领域发展。

The garment contour quantification is a prerequisite for the intelligent design and production of garments.This paper introduces the process and methods of garment contour quantification,including traditional methods and artificial intelligence-based quantification methods.It summarized the garment contour quantification method based on feature parameter recognition and mathematical formula calculations,explored the research method of contour quantification based on scanning and other technologies to construct 3D garment models,and analyzed its advantages and disadvantages.Moreover,t summarizes the garment contour quantification based on machine learning and generative network methods.Currently,most research on the quantification of clothing contours uses simple clothing samples.The study results are applied in classification and other fields,but there is relatively insufficient application in specifie numerical reasoning and judgment for each part of clothing,often remaining qualitative.The study suggests that future research can comprehensively consider various factors such as silhouette modeling,clothing styles and fabrics,and combine them with artificial itelligence technology to advance towards intelligence and automation.