Category: TECHNOLOGY AND INNOVATION
Country: Canada
An uncanny AI model decodes fabric images for knitting robots so that they can knit a complete textile without any human assistance.
By Rupendra Brahambhatt
Updated: May 04, 2025 05:57 AM EST
Imagine snapping a photo of your favorite sweater and instantly having a machine recreate it, stitch by stitch. This futuristic concept is close to becoming a reality.
A team of researchers from Canada-based Laurentian University has developed an AI model that enables knitting robots to recreate fabric patterns just by analyzing an image.
“Our paper addresses the challenge of automating knitting by converting fabric images into machine-readable instructions,” Xingyu Zheng and Mengcheng Lau, researchers from Laurentian University, told Tech Xplore.
The mechanism for fully automated knitting
Currently, if you want to recreate a knitted fabric just by looking at a photo of it, the process is slow and difficult. Someone has to carefully examine the image and manually label each stitch and pattern so that a knitting machine can understand what to do. This step is not only time-consuming but also demands a lot of expertise and precision.
It’s like translating a language one word at a time, by hand, for every single sentence. Since this entire process relies so heavily on human effort, it becomes very difficult to apply this method on a large scale. For example, if you have thousands of fabric images or want to recreate new designs quickly, this approach isn’t practical.
To tackle the challenge of translating images of knitted fabrics into precise instructions for knitting machines, the researchers devised a two-step deep learning framework that mimics the way a human expert might analyze and interpret fabric patterns.
During the first step, called the generational phase, an image of the knitted fabric is turned into a simpler, clearer version that shows only the important parts of the pattern. Think of it like turning a detailed picture into an easy-to-read sketch. This simplified image focuses on the stitches you can see on the surface, and from it, the system creates something called “front labels,” which act like a blueprint for how the fabric was made.
In the next step, the AI model uses front labels to deduce comprehensive knitting instructions, including both visible and hidden layers of stitches. These instructions are formatted in a way that knitting machines can directly understand and execute.
The researchers tested their AI model to recreate patterns for around 5000 textile samples, the results were impressive. “Our model attained an accuracy of over 97% in converting images into knitting instructions, significantly outperforming existing methods,” Haoliang Sheng and Songpu Cai, co-authors of the study, said.
Moreover, “the system effectively handled the complexity of multi-colored yarns and rare stitch types, which were major limitations in earlier approaches. In terms of applications, our method enables fully automated textile production, reducing time and labor costs,” they added.
The dark side of AI-enabled knitting
Automating the conversion of fabric images into knitting instructions can streamline production, reduce costs, and enable on-demand manufacturing. Plus, this technology could also facilitate the preservation and reproduction of ancient textile designs and patterns, which are historically and culturally important.
However, these merits may fall short to compensate for the job losses such a technology could bring. The textile industry provides employment to more than 75 million people globally, many of whom work in low-wage, labor-intensive roles such as knitting and stitching.
ABOUT THE AUTHOR
Rupendra Brahambhatt is an experienced writer, researcher, journalist, and filmmaker. With a B.Sc (Hons.) in Science and PGJMC in Mass Communications, he has been actively working with some of the most innovative brands, news agencies, digital magazines, documentary filmmakers, and nonprofits from different parts of the globe. As an author, he works with a vision to bring forward the right information and encourage a constructive mindset among the masses.
Courtesy: interestingengineering.com
Copyrights © 2025 GLOBAL TEXTILE SOURCE. All rights reserved.
