The Visual Manuscript Analysis Lab (VMA) uses advances in AI, computer vision, and pattern recognition to address practical questions arising from the study of historical manuscripts. The VMA is dedicated to the systematic visual analysis of manuscript images, delving into the multifaceted landscape of historical documents with computational precision. The VMA aims to promote a nuanced understanding of the past by applying cutting-edge technology and engaging in collaborative, interdisciplinary research.

Research in the VMA spans machine learning and computer vision, focusing on four main directions: handwriting style analysis, which examines visual features in handwritten texts to identify scribes, recognise similar styles and detect gradual changes; visual navigation of digitised manuscripts, which uses pattern analysis and multimodal learning to explore large collections; computational restoration, which applies image processing to damaged manuscripts to improve visibility and readability; and pattern analysis software tools, which automate visual and tabular analysis so scholars can gain insights more quickly and reliably.