SBIR Phase I: Using Automation to Deliver Photo-Realistic Clothing Simulations for Virtual Fittings

这个小型企业创新研究（SBIR）第一阶段项目的更广泛的影响将是展示虚拟服装创作和试穿系统的可行性。随着企业增加其电子商务业务，他们面临着一个重大挑战：电子商务的高回报率。在线购买的退货率超过实体店购买的退货率约为4比1，顾客（52-74%）认为对服装的合身性不满意是退货的主要原因。退货率降低1%，就可以使超过5000万磅的商品免于填埋，并为时尚零售商带来23 B美元的回报。这个第一阶段项目旨在开发一个复杂的过程，使用3D建模和织物模拟技术，使定制的适合和尺寸可视化之前购买。这个小型企业创新研究（SBIR）第一阶段项目将通过使用机器学习，数值模拟和3D图形渲染来展示虚拟服装创建系统的可行性，以基于（i）描述服装的图像和文本以及（ii）客户身体的最小测量集生成虚拟服装。 该奖项反映了NSF的法定使命，并通过使用基金会的智力价值和更广泛的影响审查标准进行评估，被认为值得支持。

The broader impact of this Small Business Innovation Research (SBIR) Phase I project will be to demonstrate the feasibility of the virtual garment creation and try-on system. As businesses increase their e-commerce presence, they face a major challenge: the high rate of returns in e-commerce. The return rate for online purchases exceeds that of in-store purchases by roughly 4 to 1, with customers (52-74%) citing dissatisfaction with the garments’ fit as the primary reason for returns. A reduction in returns as small as 1% could keep over 50 million pounds of goods out of the landfill and return $2.3 B to fashion retailers. This Phase I project is aimed at developing a sophisticated process using 3D modeling and fabric simulation technologies to enable customized fit and sizing visualizations prior to purchase. This Small Business Innovation Research (SBIR) Phase I project will demonstrate the feasibility of the virtual garment creation system by using machine learning, numerical simulations and 3D graphic rendering to generate virtual garments based on (i) images and text that describe the garment and (ii) a minimum set of measurements of the customer's body. This process will advance the translation of novel approaches to combining artificial intelligence and 3D representation of a deformable shape in a computationally efficient manner.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.