Tactile Embroidery Reproduction Exploiting Machine Vision for Visually Impaired Engagement

Research output: Chapter in Book/Conference proceeding with ISSN or ISBNConference contribution with ISSN or ISBNpeer-review

Abstract

Best conservation practices for historic textiles such as their display behind glass and in low light conditions to prevent damage and deterioration create significant barriers for visually impaired audiences. Alternative sensory experiences, particularly tactile exploration, have proven essential for increasing engagement with historical and cultural objects for these visitors. Leveraging state-of-the-art machine vision approaches, we present a comprehensive workflow to generate machinable 3D models across multiple materials. We evaluate these reproductions with visually impaired participants using the replica experience framework, analysing preferences across material types, tactile features, and representation techniques. Our preliminary findings demonstrate that combining contextual audio guides with tactile objects significantly enhances understanding and engagement. Notably, providing multiple material versions of the same artefact better accommodates the diverse preferences and tactile sensitivities found within the visually impaired community, suggesting material diversity should be a key consideration in developing inclusive museum experiences.
Original languageEnglish
Title of host publicationDigital Heritage International Congress 2025
PublisherThe Eurographics Association
Pages1-10
Number of pages10
ISBN (Electronic)9783038682776
ISBN (Print)9783038682776
DOIs
Publication statusPublished - 8 Sept 2025
EventDigital Heritage 2025 - Siena, Italy
Duration: 8 Sept 202513 Sept 2025
https://digitalheritage2025.unisi.it

Conference

ConferenceDigital Heritage 2025
Country/TerritoryItaly
CitySiena
Period8/09/2513/09/25
Internet address

Fingerprint

Dive into the research topics of 'Tactile Embroidery Reproduction Exploiting Machine Vision for Visually Impaired Engagement'. Together they form a unique fingerprint.

Cite this