My current research focuses on how distinctive physical objects can be connected to persistent digital identities across images, records and events — and where computational evidence should stop short of a historical claim.
Dataset and reproducibility materials supporting work on large-scale multi-view visual identity matching and an event-driven digital twin framework for handmade carpets.
Useful retrieval is not enough when the task requires continuity of identity.
Matches, extracted attributes and written records can support a claim without being equivalent to proof.
Images, structured attributes, ownership or exhibition events, condition changes and documented provenance carry different evidentiary weight.
A current direction explores AI-assisted museum documentation for handwoven carpets.
Datasets, published work and supporting research materials are added here as they become public.
Dataset and reproducibility materials supporting work on large-scale multi-view visual identity matching and an event-driven digital twin framework for handmade carpets.
Applications of Artificial Intelligence Across the Value Chain of Iranian Handwoven Carpets
Published in Dastbaf, Issue 6, this Persian-language article examines how artificial intelligence can contribute across the handmade carpet value chain — from documentation and visual analysis to digital commerce, knowledge preservation and cultural heritage.
An exploration of AI, tacit craft knowledge, provenance, authenticity, and the limits of translating handmade-carpet expertise into data.