SILKNOW · Project

AI-Powered Digital Tools That Help Textile and Fashion Companies Unlock Silk Heritage Data

digitalPrototypeTRL 5Thin data (2/5)

Imagine you have thousands of historic silk fabrics scattered across museums in six countries, each catalogued in a different language with different standards. Nobody can search across them, compare weaving patterns, or learn the old techniques. This project built a kind of "Google for silk" — an AI system that recognizes textile patterns from photos, searches across multilingual collections, and even recreates ancient weaving techniques in a virtual loom. They also 3D-printed textile designs to show how historic patterns can feed into modern fashion.

By the numbers

€525 billion

Annual turnover of EU fashion and high-end textile industries

Consortium partners across Europe

Countries represented in the consortium

Total project deliverables produced

SME partners with commercial exploitation potential

The business problem

What needed solving

Europe's fashion and high-end textile industry generates €525 billion annually, yet access to its rich silk heritage — scattered across museums in different countries and languages — remains fragmented and largely analog. Designers, manufacturers, and cultural institutions cannot efficiently search, compare, or reproduce historic weaving patterns, losing a major source of creative inspiration and premium brand storytelling. Ancient weaving techniques risk being lost entirely as specialized knowledge disappears with retiring craftspeople.

The solution

What was built

The project built a Virtual Loom that digitally models historic weaving techniques, an AI visual recognition system for automatic textile pattern identification, and multilingual semantic search tools for cross-collection discovery. They also produced 3D printed textiles with complete CAD files and fabrication documentation, plus spatio-temporal visualization tools — totaling 23 deliverables.

Audience

Who needs this

Luxury fashion houses and heritage-inspired textile brands seeking design differentiationMuseum networks and cultural institutions digitizing textile collections3D printing and digital fabrication studios exploring fashion applicationsTourism operators creating immersive cultural heritage experiencesTextile education and training institutions preserving weaving knowledge

Business applications

Who can put this to work

Fashion and High-End Textiles

mid-size

Target: Design houses and luxury textile manufacturers looking for heritage-inspired patterns

If you are a fashion brand looking to differentiate with heritage-inspired designs — this project developed a Virtual Loom and AI visual recognition system that lets you search thousands of historic silk patterns across European museum collections. Instead of months of manual research, designers can find and study weaving techniques digitally. The European fashion and high-end textile industry generates €525 billion in annual turnover, and heritage storytelling is a proven premium differentiator.

Museums and Cultural Tourism

any

Target: Museum networks and cultural tourism operators wanting to digitize textile collections

If you are a museum or tourism operator struggling to make textile collections accessible and engaging — this project built multilingual, semantically enriched search tools and visual simulations that let visitors explore silk heritage across borders. The system connects collections from 6 countries and handles multiple languages automatically. With 23 deliverables including 3D visualization tools, it offers a turnkey digitization approach for textile-focused institutions.

Creative Industries and 3D Manufacturing

SME

Target: 3D printing companies and digital fabrication studios serving fashion clients

If you are a digital fabrication studio exploring textile applications — this project demonstrated production of 3D printed textiles, complete with CAD files and documented fabrication processes. The demo deliverable includes artist designs, CAD files, and step-by-step creation documentation. This gives you a validated workflow for translating historic weaving patterns into 3D-printed fashion pieces for creative industry clients.

Frequently asked

Quick answers

What would it cost to license or use this technology?

The project was funded as a Research and Innovation Action (RIA) by the EU, coordinated by Universitat de Valencia. Licensing terms would need to be negotiated directly with the consortium. As a publicly funded project, some outputs may be available under open-access terms — check the project website at silknow.eu for current availability.

Can this scale to handle large commercial textile databases?

The system was designed to handle heterogeneous, multilingual, and multimodal databases across museum collections in 6 countries. Based on available project data, the core technology (AI visual recognition, semantic search, spatio-temporal visualization) was built for cross-collection scale. Adaptation to commercial product catalogs would require customization but the underlying architecture supports large datasets.

Who owns the intellectual property?

IP is shared among the 10 consortium partners across 6 countries (DE, ES, FR, IT, PL, SI), led by Universitat de Valencia. The consortium includes 2 industry partners and 2 SMEs who likely hold commercial exploitation rights for specific components. IP terms are governed by the EU grant agreement and the consortium's own IP arrangement.

Is this technology ready for commercial deployment?

The project ended in August 2021 and produced 23 deliverables including a working Virtual Loom and 3D printed textile demonstrations. The core AI tools (visual recognition, multilingual search) reached prototype stage with validated results. Commercial deployment would require further engineering for specific industry use cases.

How does this integrate with existing design software?

Based on available project data, the system outputs include CAD files and documented fabrication processes for 3D printed textiles. The Virtual Loom provides digital modeling of weaving techniques. Integration with commercial CAD/CAM or design software would likely require custom development, though the documented file formats suggest compatibility with standard design workflows.

What regulations or standards does this address?

The project addresses EU cultural heritage preservation requirements and digital accessibility standards. It also contributes to standards for multilingual semantic data access in cultural collections. For fashion companies, the heritage authentication capability could support provenance and origin labeling regulations.

Is there ongoing support or further development?

The project officially closed in August 2021. The project website (silknow.eu) and the consortium of 5 universities and 2 research organizations may continue maintaining certain tools. For current status and support availability, direct contact with the coordinator at Universitat de Valencia is recommended.

Consortium

Who built it

The SILKNOW consortium brings together 10 partners from 6 countries (Germany, Spain, France, Italy, Poland, Slovenia) — a strong geographic spread across Europe's key textile heritage regions. The team is research-heavy with 5 universities and 2 research organizations, reflecting the project's academic orientation. Only 2 industry partners (both SMEs) participated, giving a 20% industry ratio, which is low for commercial readiness. The coordinator, Universitat de Valencia in Spain, is a higher education institution. For a business looking to adopt these tools, the limited industry involvement means you would likely need to invest in further productization, but the academic depth ensures the underlying AI and heritage modeling is rigorous.

UNIVERSITAT DE VALENCIACoordinator · ES
UNIVERSITE LUMIERE LYON 2thirdparty · FR
INSTITUT JOZEF STEFANparticipant · SI
UNIVERSITA DEGLI STUDI DI PALERMOparticipant · IT
EURECOM GIEparticipant · FR
CENTRE NATIONAL DE LA RECHERCHE SCIENTIFIQUE CNRSparticipant · FR
GOTTFRIED WILHELM LEIBNIZ UNIVERSITAET HANNOVERparticipant · DE

How to reach the team

Contact the research team at Universitat de Valencia (Spain) through SciTransfer for a qualified introduction.

Order a report on this consortium See UNIVERSITAT DE VALENCIA profile →

Next steps

Talk to the team behind this work.

Want to explore how AI-powered textile heritage tools can give your fashion or museum business a competitive edge? SciTransfer can arrange a direct introduction to the SILKNOW research team and help you assess commercial licensing options.

Order a report on this topic More in Digital & ICT