Overview

Cultural heritage institutions are increasingly using 3D digital preservation techniques to preserve fragile museum artefacts and to develop virtual online collections. Such 3D representations improve access to physical collections and provide precise digital surrogates for scholarly analysis. As 3D data acquisition tools become more affordable and available, the amount of 3D data requiring storage, indexing and search services will grow dramatically. Because high quality metadata is prohibitively expensive, museums are keen to explore how they might exploit community tagging and annotation services.

Such services enable users to easily attach information, comments and keywords to 3D objects and their sub-parts and hence, add significant contextual value to the collection, enhance discovery and enable much richer online exhibitions. Annotation services also facilitate collaboration and knowledge exchange between geographically dispersed communities of curators, scholars and Indigenous owners. Moreover, if the annotations are drawn from pre-defined ontologies, they provide machine-understandable semantics that enable further reasoning about the object and the derivation of new knowledge.

The aim of this project is to develop simple, semantic annotation services for 3D digital objects (and their sub-parts) that will facilitate the discovery, capture, inferencing and exchange of valuable cultural heritage knowledge.

• Annotations of 3D models for efficient indexing, search and retrieval of 3D artefacts from large-scale museum digital libraries based on HTML5/WebGL standards and Semantic Web technologies.
• Semantic annotations that are drawn from an ontology to suggest useful terms.
• Annotation of meaningful parts or features of a 3D model supported by an ease to use interface.
• Generate and display a high visual quality presentation in a restrictive web environment, balancing between visual quality, download size and performances.

Personnel

Principal Investigator: Professor Jane Hunter

Jane Hunter is Director of the eResearch Lab at the University of Queensland where she leads a team of Post-docs, PhD students and software engineers developing innovative software services for managing digital collections for the cultural, education and scientific domains. Her expertise is in semantic markup and interoperable annotations to facilitate the virtual integration of distributed online collections via the Semantic Web and Linked Open Data. She is currently a CI on: the Mellon-funded Open Annotations Collaboration project; the ARC-funded 20th century Painting Conservation Knowledgebase project; the Aus-e-Lit project; and the 'Semantic Annotations for 3D Museum Artefacts' project. She is also the manager of 6 ANDS (Australian National Data Services) based at the University of Qld. She has published over 90 peer-reviewed publications, many on the application of semantic web technologies to collections and datasets in libraries, archives and museums. She was the chair of JCDL/ICADL 2010 and is also deputy chair of the Australian Academy of Sciences Committee for Data in Science and represented Australia at the recent CODATA conference.Collaborators in cultural heritage domain include:

• The Smithsonian National Museum of American Indian
• Library of Congress
• National Library of Australia National Museum
• Centre for Cultural Materials Conservation
• Queensland Art Gallery
• Museum and State Library

Research assistance: Mr Chih-Hao Yu (David Yu)

David Yu, a dedicated research associate with post-graduate experience in software engineering in 2D/3D image processing, semantic web and multimedia digital libraries that was employed to work on this project. David had completed his Multimedia Design Honours from University of Queensland in 2008. Currently, working on this project for his Phd degree, under Prof Hunter's supervision.

The major role of research assistant involves working closely with museum staff to generate the 3D models and implement and evaluate the semantic annotation services.

Sponsors

3D Gallery / 3D Annotation prototype v2.0

This project had a repository of 3D objects that were scanned from artifacts from UQ Antiquties Museum and UQ Anthropology Museum, using Konica Minolta Vivid-9i non-contact laser scanner. 3DSA annotation service is developed based on HTML5 and WebGL standards, which provides functionalities such as displaying, segmenting and tagging/annotating 3D museum object - without any plugin installation (still requires WebGL compatible browsers).

Below is a list of features:

• Presentation of high quality 3D museum objects in full colours.
  • Museum object with good visual quality balanced with performance - 65,000 polygons.
  • Museum object with optimum visual quality - 500,000 polygons, almost 8X higher than the original.
    (Requires a strong graphical computer and fast internet.)
• Tagging support for both resolutions of object, annotations are interoperable between the two versions.
• Point annotation - tag to a particular location on the 3D object.
• Segment/surface annotation - tag to a user-defined segment or surface on the 3D object.
• Intuitive user interface
  • Large 3D display area.
  • Customisable interface.
  • WYSIWYG interface for tagging.

This service is currently supported by WebGL compatible browsers.

• Mozilla Firefox - version 4 above
• Google Chrome - version 9 above

If the recommended browsers failed to run 3DSA system, please visit the "Troubleshooting" section.

Collaborators

We also envisage that this project will facilitate collaborations between the University of Queensland and the following research groups within Microsoft Research Asia:

• The TextureMontage group from Internet Graphics - to improve the texture mapping of the annotated 2D images onto the surface of the 3D objects.
• The Visual Computing group - to improve 3D object recognition through a combination of machine learning, semantic annotation and automatic (SWRL) inferencing rules.

Related Works