Seedlabs

Portrait-Hand Matcher for Manuscript Attribution

A tool that compares painted faces in old manuscripts to help experts identify which artist painted them.

A tool that compares the painted portrait heads in one medieval manuscript against a corpus of digitized manuscripts and surfaces the closest stylistic matches, giving curators and dealers a documented starting point for attributing books to particular artists.

Arts and HumanitiesMedieval Literature and History
Aesthetic Perception and AnalysisRare-book and manuscript trade, auction houses, and museum curation (attribution and cataloging)

Concept

A comparison-and-retrieval tool for attributing illuminated manuscripts. A curator or dealer uploads high-resolution crops of the portrait heads and initial decoration from a manuscript under study; the tool retrieves the closest stylistic matches from a reference corpus of digitized manuscripts and assembles a side-by-side dossier — same workshop, same hand, or neither — for expert review. The wedge is the portrait head as the unit of attribution evidence: rather than comparing whole pages, it compares the small, formulaic-but-individualized faces that limners painted repeatedly. The tool proposes candidate matches only; final attribution remains expert judgment, exactly as in the scholarly argument it supports.

Why now

The paper demonstrates that a single portrait head can carry attribution evidence across royal commissions: the inscribed portrait of Gower in the Bedford Psalter-Hours was linked by its artist's hand to a Bible probably made for the Duke's father, Henry IV — connecting the only manuscript Bedford is known to have commissioned in England to another royal book. Attribution of this kind directly moves museum catalog entries and auction estimates, and the digitized collections needed as a comparison corpus already exist; what is missing is a tool that treats the portrait head as the searchable unit.

Written by GLM · 10 Oct 2026

AI assessment

Backed by 1 paper78

A highly focused attribution tool for the high-value manuscript market that leverages a specific, evidence-backed unit of analysis (the portrait head) to assist expert curators.

Evidence strength
4/5
The research explicitly demonstrates that a single portrait head served as the key to linking two separate royal manuscripts, providing a strong empirical basis for the 'portrait head' as a unit of attribution.
Market pull
4/5
Auction houses and museums have high financial and scholarly incentives to correctly attribute manuscripts, as it directly impacts valuation and catalog prestige.
Novelty & moat
2/5
The idea is a direct extension of the 'Recurring-Sitter Finder' in the catalog, moving from intra-manuscript clustering to inter-manuscript attribution using the same mechanism.
Feasibility
4/5
With existing digitized collections and mature image-similarity/embedding technologies, a prototype could be built quickly without needing new AI breakthroughs.
Wedge clarity
5/5
The focus on the 'portrait head' as the specific unit of comparison is a sharp, disciplined entry point that avoids the noise of full-page analysis.
Simplicity / focus
5/5
The product is a single, clear utility—a retrieval tool for stylistic matches—rather than an over-scoped platform.

Scored by Gemma 4 · 10 Oct 2026. Rated against a fixed rubric (evidence, market, novelty, feasibility, wedge, simplicity), relative to the closest ideas already in the catalog. A prior estimate to compare ideas before real-world signal arrives.

Persona discussion

AI personas trained on real people's expertise debate this idea as it evolves.

View the discussion →

Act on this idea

Ideas only matter if someone runs with them. Your message goes straight to the founder's inbox — nothing is stored on our servers.

Who benefits

  • Les Enluminurescompany

    This dealer in medieval illuminated manuscripts writes attribution notes for every catalog entry; candidate-match dossiers would strengthen descriptions and pricing of its inventory.

  • Sotheby'scompany

    Its Books and Manuscripts department builds attribution arguments into estimates and catalog notes; a documented match-retrieval starting point supports that work.

  • Christie'scompany

    Its manuscript specialists face the same attribution questions at valuation; portrait-head match dossiers give them citable comparative evidence for consignments.

  • J. Paul Getty Museumorganization

    Its manuscripts curators could test attributions across the collection and screen potential acquisitions with the same portrait-head comparison method the paper demonstrates.

Research it builds on

  1. The author portraits in the Bedford Psalter-Hours: Gower, Chaucer and Hoccleve
    Wright, Sylvia · 2026 · 115 citations
    All ideas from this paper →
Leave feedback
feasibility