This website uses cookies to ensure you get the best experience. By clicking or navigating the site you agree to allow our collection of information through cookies. Check our Privacy policy.

2 minutes to read Posted on Thursday December 19, 2024

Updated on Thursday December 19, 2024

Exploring the importance of digital media authenticity in cultural heritage

What role does authentic, high-quality digitisation play in understanding and preserving our cultural past? How are developments in artificial intelligence affecting authentic digitisation? A recent event from the EUreka3D project discussed the topic - discover the learnings and reflections.

By Frederik Temmermans and Jolan Wuyts

People inspecting and tasting tea
Title:
The Preparation of Tea in Japan. Inspecting and tasting tea
Date:
1902
Institution:
Rijksmuseum
Country:
Netherlands

In a recent webinar, organised by the EUreka3D project consortium and International Council on Archives (ICA) on digital media authenticity, Frederik Temmermans from VUB/ imec (a project partner of EUreka3D) shared his expertise on digital media authenticity, particularly in preserving cultural heritage. This talk emphasised the importance of authentic, high-quality digitisation in understanding and preserving our cultural past. Read on to discover what was discussed.

The challenge of digital media authenticity

There are clear challenges in capturing and storing authentic digital representations of physical objects. In this context, a digital media asset is considered authentic if it represents a true representation of the original artefact and is unaltered since its inception. Artworks, for example, can vary widely in appearance when digitised under different conditions. Differences in lighting, camera equipment, and even subtle environmental factors like temperature can all affect the outcome. These inconsistencies make it difficult to assess artefacts scientifically and to compare digital reproductions over time.

These problems become apparent when searching the web for iconic artworks such as Leonardo da Vinci’s Mona Lisa or Pieter Bruegel’s Tower of Babel. Discrepancies in colour, contrast, and even proportions can complicate the viewer’s understanding of the original piece. Copies and parodies are not always easy to identify.

EUreka3D and the importance of paradata

The EUreka3D project aimed to address these concerns by promoting the consistent capture and preservation of what is known as ‘paradata’—detailed information describing the conditions and settings of each digitisation event. This paradata includes information about the camera settings, lighting setup, environmental conditions and the specific equipment used. Capturing these details ensures that digital representations are more reliable and comparable over time.

Recording this information allows historians, scientists and the public to trace a digital record back to its source conditions. In the EUreka3D framework, preserving paradata alongside digital images is crucial for ensuring any future analysis of these digital records while accounting for variations resulting from variations in the capture conditions.

This data-rich approach is especially important as digital reproductions increasingly serve as proxies for physical access to artefacts. By preserving paradata, EUreka3D aimed to bridge the gap between static digital captures and the dynamic, evolving reality of historical objects. For instance, through the accurate capture of paradata, you could theoretically track changes across multiple digitised versions of the same artefact, recognising alterations due to environmental factors, restoration efforts, or technological advances.

Digital media manipulation and the role of AI

The growing role of artificial intelligence (AI) in digital media, both as a tool for enhancing images and as a source of potential manipulation, is an important factor in this discussion. AI driven tools are available to improve visual quality, for example by denoising, colorising or increasing the image resolution.

However, such tools can introduce new information not present in the original work. While these developments provide valuable new possibilities, they further blur the line between authentic representations and digital alterations. The fact that those tools become more widely accessible poses challenges for distinguishing real artefacts from manipulated versions. As an illustration, the example below shows an image of Da Vinci’s Mona Lisa where the background was extended using the Generative Expand AI tool built in Adobe Photoshop.

The Mona Lisa with an extended background
Title:
Mona Lisa with an extended background, generated by Photoshop’s Generative Expand feature
Creator:
Leonard Da Vinci; Frederik Temmermans
Date:
2024
The Mona Lisa with an extended background

AI-based detection methods are commonly used to identify generated and manipulated content. While such methods have lately become impressively accurate, the problem becomes an adversarial battle between AI tools. As detection tools improve, so do the methods used to create the content, leading to an ongoing game of ‘cat and mouse’. As a sector, we have to work towards a future where provenance of digital media is precisely documented, starting from the creation of the media and continuing over the entire lifecycle. EUreka3D exemplifies this proactive approach via the detailed documentation of paradata. To further safeguard the authenticity of digital heritage, paradata and metadata need to be securely linked to the media content by adopting cryptographic methods.

Looking ahead: interoperability and standardisation

To make this vision for the future a reality, interoperability and standardisation will be imperative. The JPEG Trust standard (ISO/IEC 21617) aims to create a standardised framework for embedding and securing provenance data in digital media files. This initiative seeks to align efforts across organisations and technologies, allowing institutions worldwide to share and verify authentic cultural records more effectively.

EUreka3D’s emphasis on both paradata and metadata as fundamental elements of digital preservation points toward a more transparent digital future, where provenance data ensures that cultural heritage remains reliable and accessible for generations to come.

Find out more

If you would like to find out more about EUreka3D and its achievements, find out how the project enabled the creation and reuse of 3D heritage. You can also watch a recording of Frederik’s talk on the EUreka3D website.

top