AI for Historical Textiles: Recognizing Interlaces and…

AI for Historical Textiles: Recognizing Interlaces and Conservation

Textiles preserve technologies, trade, and aesthetics. AI automates the analysis of interlacements, defects, and material degradation, aiding in conservation and reconstruction planning.

Structural Recognition (Serape, Atlas, Canvas)

This focuses on identifying the specific weave structures within historical textiles – such as serape, atlas, or canvas patterns.

live demo · related simulation● LIVE

Provenance Classification by Patterns

Recommendations for conservation and storage are derived from pattern analysis, considering scan quality and the variability of manual techniques.

Explainable Decisions for Curators

The infrastructure and data used support informed decisions, incorporating macro-photography, microscopy, and multispectral imaging technologies.

Frequently asked questions

What are metadata schemas and annotation tools used in this project?

Metadata schemas and annotation tools facilitate the structured organization and labeling of textile data, ensuring accuracy and consistency.

How are reference datasets collected and used to train the classification model?

Reference datasets – carefully labeled examples of textile patterns – are gathered and utilized to train a machine learning classifier, improving its accuracy in identifying different weaves.

Can a prototype be developed to predict textile degradation?

A preliminary prototype can be created to forecast potential degradation based on material properties and environmental factors, allowing for proactive conservation strategies.

How will this AI system integrate with existing collection management systems?

Integration with current collection management systems is a key goal, streamlining workflows and providing curators with valuable data insights.

▶ Try it live

Everything above runs in your browser — open Hash Function Avalanche Visualizer and change the parameters while it is running. Nothing is installed, nothing is uploaded, the whole model lives in one tab.

▶ Open Hash Function Avalanche Visualizer simulation