SBIR Phase I: Weft: Simulation Software for Textile Design and Manufacturing

The broader impact/commercial potential of this Small Business Innovation Research (SBIR) Phase I project is to revolutionize the way that woven textiles are designed and manufactured. The textile industry is currently organized around fixed catalogs of fabrics, with significant barriers to the introduction of new products, including limited access to industrial equipment. The project will create Weft, a new web-based design system and marketplace organized around a user-friendly online tool for designing and previewing woven textiles. Anyone can log in, design fabric based on their own graphics and images, order it to be produced in any quantity, and offer it for sale in the Weft marketplace. Weft's fluid digital manufacturing flow can spark a revolution analogous to desktop publishing: any designer with an idea can rapidly and easily turn it into fabric. Smaller projects become possible; larger projects are done faster and better; and customization becomes commonplace. Furthermore, reimagining textile production can revitalize the US textile industry, shifting from old-fashioned mass production to a nimble, customized market that rewards manufacturers and designers with flexibility, technical skills, and proximity to the end-user. This Small Business Innovation Research (SBIR) Phase I project will commercialize NSF-funded research on realistic rendering of textiles. A core technology of Weft is its unique visualization technology, which lets users see the exact appearance of their designs in advance, with full confidence that they will receive fabric that looks just like the visualization they saw on their screen. Research has developed a pipeline for fabric rendering based on scanning the detailed arrangement of fibers in small samples of cloth, then using optical simulation to compute the flow of light through fabric woven in any pattern and with any combination of colors, thereby predicting the exact appearance. The simulation results are realistic and accurate, but the technology has only been demonstrated for one particular type of fabric, and computation times are too long for interactive use. As part of developing the technology for deployment in a web application available to the public, the project will adapt the existing technology to (a) work for a variety of fabrics beyond the one type demonstrated in research, and (b) produce results, under the particular conditions needed for the design tool, at interactive speed.