The Intricate History of Textiles and Computing
Intricate historic interweaving of textile manufacturing and digital technologies opens unlimited transformational opportunities at the dawn of the new era of artificial intelligence (AI), which is reshaping industries and the fabric of human societies. While fashion and textile industries already leverage AI-powered tools for real-to-virtual transformation of products and processes, we make the case that AI can and should play a key role in enhancing virtual-to-real product transformation via generative design of textiles for manufacture.
The connection between textiles and computing runs deep, tracing back to the early days of the digital revolution. Innovations in textile weaving, such as the use of binary and nonbinary code for information storage and exchange in the Jacquard loom, laid the foundation for modern computing. Even the first nonvolatile computer memory was physically woven by textile workers into fabric-like patterns. It is perhaps symbolic that today’s AI language models, such as ChatGPT, tend to overuse textile-related words like “weaving” and “tapestry” – a testament to the intricate relationship between the language of words and the language of textile patterns.
Bridging the Gap: Generative AI for Textile Design and Manufacture
Despite the long-standing connection between textiles and computing, the role of AI in today’s fashion and textile industries has largely been limited to digital design and production quality control. The biggest impacts of the AI revolution in textiles have been in the digital domain, such as managing supply chains, enhancing shopping experiences, and digitizing textile collections for heritage preservation.
We believe that AI can and should play a more significant role in enhancing “virtual-to-real” product transformation via the generative design of textiles for manufacture. AI capabilities in this field remain largely untapped and need to be developed to engineer enhanced mechanical properties, transform textile manufacturing processes, and enable smart textiles applications.
Bobbin Lace: A Challenging Example
To illustrate our proposed generative AI-enabled pipeline, we focus on bobbin lace as an intricate, challenging example of an endangered handicraft important to textile heritage. Lacemaking draws from elements of weaving, embroidering/sewing, and knitting, all current methods of mass textile manufacture, but adds the challenge of holes and negative space in the fabric, forming an intricate pattern.
Relative to more conventional woven or knitted textiles, the open net structure of lace textiles provides additional degrees of freedom in tensile properties engineering, which can be leveraged for modern applications in wearable, medical, industrial, and geo textiles. AI-generated lace patterns can be optimized for aesthetic appeal, cultural relevance, elasticity, tensile strength, Poisson ratio, and other mechanical characteristics, as well as for the integration of conductive threads and electronic components.
The Evolving Landscape of Generative AI for Fashion and Textiles
The fashion and textile industries have already witnessed the impact of generative AI models, such as Generative Adversarial Networks (GANs), Variational Autoencoders (VAEs), and Diffusion Models, in creating new designs. These deep learning techniques encode stylistic information into a lower-dimensional latent space, allowing the generation of new patterns and designs by sampling from the learned probability distributions.
However, the current phase of AI-generated content typically takes the form of pieced-together images that are not yet manufacturable. The challenge lies in creating generative AI models capable of producing patterns that are complete, physically possible to be made, and encoded to be produced by hand or machine.
Designing for Manufacturability: A Three-Stage Process
To maintain cohesion between the generation and instruction-translation models, we identify three critical stages in the AI-enabled textile design-for-manufacture process:
-
Attribute-Specific Pattern Generation: The attributes may comprise aesthetic as well as textural, mechanical, or thermal features.
-
Process-Specific Instructions Encoding: The process-specific instructions can take many different forms, depending on the choice of the manufacturing process (i.e., weaving, knitting, bobbin lacemaking, three-dimensional printing, or embroidery).
-
Physical Fabrication: Integrating the generated patterns and process-specific instructions to physically produce the textile.
Knitting Encoding Techniques
Modern knitting is a fully computerized textile construction technique, which creates patterns from interlacing yarn loops comprising various stitch types (e.g., knit, purl, tuck, flow, etc.) used as pixels. Punch cards, originally created for the Jacquard loom, have provided a way of encoding lace in a binary way while retaining the ability to make intricate designs.
Researchers have explored various deep learning approaches to establish a design-for-manufacture pipeline, including mapping real and synthetic images to knit instructions and generating new patterns using long short-term memory architectures.
Bobbin Lace Encoding Techniques
Bobbin lace is made by braiding and twisting filaments or yarns, which are wound on multiple bobbins. Simple movements of the bobbins (e.g., twists and crosses) create stitches according to a predefined pattern. Handcrafted bobbin lace instruction typically includes both an encoded sequence of stitches and a visual model of pricking patterns.
An alternative lace encoding approach has been recently proposed to represent these patterns as graphs, which allows effectively integrating bobbin lace patterns as quantifiable data representations. Graph neural networks can then be used to generate not only realistic textile graphs that reflect these relationships but also reproducible instructions to create them.
3D Printing and Embroidery Encoding
The adaptation of three-dimensional (3D) printing techniques for lace manufacture faces different challenges. While a sophisticated and flexible system of coding 3D patterns is well developed, reaching the same level of material flexibility, aesthetics, and production rate as those achieved with knitted or bobbin lace remains a challenge.
On the other hand, digital embroidery techniques offer an alternative approach to constructing 3D interlocking lace patterns. The process involves repeated patterns produced by a computer-controlled embroidery machine in layers, which can build up new interlocking mechanisms.
Engineering Lace Attributes for Modern Applications
While the aesthetic properties of lace structures may be driving their consumer appeal, it is the mechanical properties that play a critical role in determining the suitability of different patterns for various applications. Key attributes include tensile strength, elasticity, dimensional stability, fineness, and texture.
Generative AI is expected to expand the possibilities of incorporating these attributes and nontraditional materials in lacemaking practice, elevating the craft to be of research significance and bringing functional and aesthetic value to modern textiles.
Characterizing Lace Mechanical Properties
Experimental evaluation of the tensile properties of different lace patterns can provide valuable data to inform the training of generative AI models. Parameters such as ultimate strength, Young’s modulus, and characteristic transition strain, as well as the interplay between solid threads and open spaces (porosity), can be quantified and used to condition the AI-generated patterns.
Computer vision techniques can also help derive detailed strain maps from high-resolution images of lace under deformation, serving as surrogate models for the generative AI and guiding the optimization of mechanical properties.
Balancing Tradition and Innovation: The Path Forward
The integration of generative AI in textile engineering presents both challenges and opportunities. While it can streamline the design and manufacturing processes, it is essential to maintain the human element and the cultural significance of traditional textile crafts.
Educators and institutions must advocate for progressive change, addressing concerns around the ethical use of AI and demonstrating the complementary nature of these tools in enhancing, rather than replacing, the personal engagement and mentorship that are central to effective teaching.
By embracing AI as a powerful ally in our mission to educate, inspire, and prepare, we can redefine the educational landscape and create a legacy that resonates with the future we are helping to build. The future of textile engineering lies in the seamless integration of generative AI and the timeless art of craftsmanship, blending tradition and innovation to unlock new possibilities.
