Art Speaks Maths, Maths Speaks Art

Year: 2020 Authors: N. Leone; S. Parisotto; K. Targonska-Hadzibabic; S. Bucklow; A. Launaro; S. Reynolds; C.-B. Schönlieb

Core claim

Maths can effectively serve cultural heritage only when arts and humanities experts first define the questions, data, and validation criteria.

Topics

cultural heritage computing, cross-disciplinary collaboration, image analysis, classification, virtual restoration

Domains

machine learning, content-based image retrieval, k-nearest neighbours, image inpainting, variational methods, painting conservation, art history, archaeology

Methods

deep neural network segmentation, feature-based similarity search, mathematical classification, digital reconstruction, iterative user feedback

Media

painting cross-sections, Roman pottery profiles, digitized manuscripts, archival images, software toolkits

Source status

This page publishes metadata and extracted analytical signals only. Raw PDF and full OCR text are kept local for now.