RSS feed for this blog

Mapping VRA Core 4 to VisualArtwork

By Paul Watson on . - A series of 3 posts: post 2 of 3.

What is VRA Core?

VRA Core is a data standard for the description of works of visual culture as well as the images that document them. Works of visual culture can include objects or events such as paintings, drawings, sculpture, architecture, photographs, as well as book, decorative, and performance art. It is an internationally recognized metadata standard that is used both as a standalone format, and as an approved extension schema to METS for objects that contain cultural heritage resources.

An Introduction to VRA Core 4 (PDF).

If you’re an individual artist or small gallery then you’ve probably never heard of VRA Core 4, and to be honest you probably don’t need to care about it - you won't be using it on your website and there’s probably no point in using it on your website. It’s an XML schema that is used to structure and share metadata about visual artwork that is used by large cultural institutions such as museums.

If you’re interested in learning about it (and in other metadata schemas for visual artwork) then I would highly recommend Stephen J. Miller’s 2011 book Metadata for Digital Collections: a how-to-do-it manual (Facet Publishing/Neal-Schuman; ISBN: 978-1-85604-771-5; companion website).

I’m going to borrow the author’s sample VRA Core 4 XML file from chapter 8 (pages 222–224) to illustrate how to map VRA Core 4 to This is a common sample file provided by the Visual Resources Association, and can be viewed at

The artwork and the image of the artwork

VRA Core 4 has two main elements: work and image. The former holds the metadata about the actual piece of artwork (e.g. the painting), while the latter holds metadata about the digital scan/photograph of that piece of artwork (e.g. the JPG representation of the painting):

<?xml version="1.0" encoding="UTF-8"?>
<vra xmlns=""
    <work id="w_26" source="Core 4 Sample Database (VCat)" refid="26">
    <!-- metadata about the actual artwork -->
    <image id="i_129"
        refid="129" source="VRA Core Oversight Committee, Core 4 Sample Records">
    <!-- metadata about the SCAN of the artwork -->
</vra> provides two ways to mark-up an image of a piece of visual artwork (or indeed anything) - as a simple URL, typically in the src attribute of an <img> element, or as an ImageObject. A simple URL is by far the most common use, but for this task we need to use the ImageObject. Let’s set up the basic structure for the VisualArtwork type with an ImageObject:

<div vocab="" typeof="VisualArtwork">
    <!-- data about the piece of artwork goes here -->

    <div property="image" typeof="ImageObject">
        <!-- data about the *scan* of artwork go here -->


Material and Surface

The example XML file unfortunately does not split the materials into individual elements or distinguish between material and surface, but such clarity is possible using VRA Core 4 as shown in this example (PDF, bottom of page 2):

  <display>oil on canvas</display>
  <notes source="Art Bulletin, v.87, no. 1 (March 2005)">Medium originally thought to be tempera. Oil medium discovered in tests at Uffizi in 2003</notes>
  <material type="medium" vocab="AAT" refid="300015050">oil paint</material>
  <material type="support" vocab="AAT" refid="300014078">canvas</material>

For details about implementing a Linked Open Data approach to link the Getty AAT identifiers to URIs see my next post Getty AAT Linked Open Data in’s VisualArtwork.

The full mark-up with explanatory comments

<div vocab="" typeof="VisualArtwork">
    <!-- The <display> element from <titleSet> -->
    <h1 property="name">
        Autumn - On the Hudson River
    <!-- The <name> element from <agentSet> -->
    <p property="creator" typeof="Person">
        Artist: <span property="name">
            Cropsey, Jasper Francis
    <!-- The <display> element from <descriptionSet> -->
    <div property="description">
            This monumental view of the Hudson River Valley was painted from 
            memory in the artist's London studio. Cropsey adopted a high vantage 
            point, looking southeast toward the distant Hudson River and the 
            flank of Storm King Mountain. A small stream leads from the 
            foreground, where three hunters and their dogs gaze into the sunlight. 
            All along the meandering tributary there are signs of man's peaceful 
            coexistence with nature: a small log cabin, grazing sheep, children 
            playing on a bridge, and cows standing placidly in the water. Here, 
            man neither conquers nor is subservient to nature; both coexist 
            harmoniously. In fact, the landscape is depicted as a ready arena 
            for further agricultural expansion. Cropsey's painting is a 
            celebration of American nationalism.
        <!-- The <display> element from <materialSet> -->
        <!-- These should be available as individual elements in your XML -->
            <span property="artMedium">oil paint</span> on 
            <span property="artworkSurface">canvas</span>
        <!-- The <measurements> elements from the <measurementsSet> -->
            <span property="height">151.8 cm</span> × 
            <span property="width">274.9 cm</span>.
        <!-- The <worktype> element from the <worktypeSet> -->
        <!-- For further information about the URL please see -->
        <!-- -->
            <span property="artform" 
                oil painting (visual work)
        <!-- The delimited contents of the <display> element from the <subjectSet> -->
            <span property="keywords">agriculture</span>, 
            <span property="keywords">animals</span>, 
            <span property="keywords">domestic life</span>, 
            <span property="keywords">genre</span>, 
            <span property="keywords">landscapes</span>, 
            <span property="keywords">Hunting in art</span>, 
            <span property="keywords">Rivers</span>,
            <span property="keywords">Hudson River</span>,
            <span property="keywords">autumn foliage</span>,
            <span property="keywords">trees</span>,
            <span property="keywords">leaves</span>,
            <span property="keywords">Fall</span>.

    <div property="image" typeof="ImageObject">
        <!-- metadata about the *scan* of the artwork goes here -->
            <img property="image" src="" />
                <!-- The <display> element from the <titleSet> -->
                <p property="caption">Overall view without frame</p>
                <!-- The <display> element from the <rightsSet> -->
                <!-- If the scan (not the artwork) were copyrighted then -->
                <!-- the copyrightHolder and copyrightYear properties would be used -->



Bibliography and further reading:

Miller, S. (2011). Metadata for digital collections: A how-to-do-it manual. New York, NY: Neal-Schuman.

VRA Core Support Pages:

VRA Core ListServ:

Other posts in this series - A series of 3 posts: post 2 of 3:

  1.’s VisualArtwork launched
  2. Mapping existing data stored in VRA Core 4 format to’s VisualArtwork
  3. Getty AAT Linked Open Data in’s VisualArtwork