Silk meets Science: European Heritage Award went to EU research project SILKNOW

Silk has been synonymous with craftsmanship, beauty and luxury for hundreds of years. The silk trade was also an important factor for progress in Europe, enabling the exchange of ideas and innovations. For instance, long before modern computers could be thought of, punch cards were first used in Jacquard silk looms. Today, however, silk textiles and artisanal weaving techniques are a highly endangered heritage. Therefore, SILKNOW was dedicated to the preservation and dissemination of knowledge related to the history of silk production in Europe - also using AI.

Preservation of ancient weaving techniques

During the project, which  had been funded from 2018 to 2021 under the EU framework programme for research and innovation “Horizon 2020”, new methods based on state-of-the-art procedures from IT and communications technology were developed to preserve the intangible heritage of ancient weaving techniques. This was achieved by the collaboration of a multidisciplinary team comprising experts in the fields of cultural history and textile conservation, computer science and artificial intelligence for image and text analysis, as well as small enterprises involved in 3D printing and textile fabrication. The Institute for Photogrammetry and Geoinformation at Leibniz Universität Hannover (LUH) was one of the main players in this international project, with two main contributions related to image analysis on the basis of AI. In addition to Leibniz University Hannover, institutions in Spain, Italy, France, Slovenia and Poland participated in the consortium led by the University of Valencia.

"What was special about this project was its interdisciplinary nature. It took a while for the participants to learn each other’s scientific language. At the beginning, this was a challenge, but as soon as this had been accomplished, we could make progress in the field of cultural heritage preservation that could not have been achieved by any of the involved groups alone.“, says apl. Prof. Franz Rottensteiner, the project leader at LUH. "We utilize methods from artificial intelligence to derive information about the depicted textiles - such as time and place of manufacture - from images of historical silk textiles", he continues. Thanks to these methods, information about silk textiles that was not originally available in the digital data sets is automatically derived from images. Mareike Dorozynski, a PhD student of Dr. Rottensteiner who developed these methods, adds, "Another method developed in Hanover enables the search for data sets based on the similarities of images. In this way, the textiles that are most similar to the image can be found. This procedure then allows access to the information linked to these textiles".

Applications for the Future

"An artificial neural network uses existing data to learn, for example, the visual characteristics of silk textiles from a certain period and then applies this model to other images of silk textiles. Thanks to this automatic classification, collections containing incomplete or non-standardized data can be completed. In future, unified information and extensive search functions will make it easier for cultural historians to research different collections", says Dr. Rottensteiner. Within the scope of SILKNOW, a knowledge graph featuring nearly 40,000 entries on artefacts made of silk with images and further information could be used for training. It is now possible to automatically derive information such as the time and place of production from images available in the platform ADASilk.

The project also produced a "Virtual Loom", developed in Valencia, in which different weaving techniques - often only known to those in the trade - are visualized and preserved for future generations. A further result of SILKNOW includes a multilingual thesaurus, which contributes to standardized use of terms when analyzing and describing silk textiles, and enables terms and phrases to be searched in different languages.

Knowledge Transfer

No wonder this project has been awarded several prices. Within the European Heritage Award / Europa Nostra Award, SILKNOW also gained the "Grand Prix for Innovation". The "Grand Prix" honours outstanding projects from all the European Heritage Award / Europa Nostra Award-winning projects in up to five categories, innovation being one of them. Furthermore, SILKNOW was included in the European Commission's Innovation Radar - an initiative by the European Commission to identify innovation with great potential in EU-funded research and innovation projects: SILKNOW improves the understanding of EU heritage and its rich diversity, applying next-generation ICT research to the needs of various users like museums, education, tourism, creative industries or media. It also preserves an intangible heritage like the ancient weaving techniques for next generations.

Last but not least, anyone can use SILKNOW. Thanks to the Virtual Loom, one can visualise finished designs before they are woven - colours, yarns, weaves and techniques can be chosen to experiment with new designs and materials. ADASilk even goes further by accessing silk heritage to replicate or reinvent past designs and motifs. Ancient techniques for tomorrow's applications!

„With the help of the collaborative effort, SILKNOW has produced a basis for analyses by cultural historians working on the history of the European silk production, but it has also set standards for digitization of cultural heritage that can be relevant for other applications,” Dr. Rottensteiner says. “Additionally, the virtual loom can not only be used to preserve information about weaving techniques; in combination with a 3D printer that can print on textiles, it allows the use of this information by the fashion industry,” he continues. For example, a collection of fashion items based on the virtual loom was produced by the Polish fashion designer Patriyk Wojciechowski. It should have been presented at the Berlin fashion week; due to Covid, the catwalk had to take place in Warsaw and could only be presented online (on YouTube). The success of SILKNOW has motivated the consortium members to submit new proposals for EU projects related to cultural heritage.