Author: Sanjay Goel
The other seven articles in this series give an overview of how the field of cultural heritage is applying different computing technologies – Virtual Reality (VR), Augmented Reality (AR), Internet of Things (IoT), Text Mining, Natural Language Processing (NLP), Geographical Information Systems (GIS), Computer Vision, Graphics, Image Processing, Robotics, and Digital Audio/Video Processing. These computing technologies are redefining the opportunities, scope, workflows, and engagement in activities related to the discovery, documentation, preservation, conservation, restoration, education, management, or enrichment of cultural heritage.
___________________________________________________________________________
In today’s era of rapid technological advancements, disciplines like cultural studies, history, archaeology, art history, museology, etc., continue to be vital for understanding human history, preserving cultural artifacts, enrichment of society, and fostering a sense of shared heritage. They contribute to maintaining cultural continuity, shaping perspectives, promoting dialogue, addressing global challenges, and stimulating creativity. In essence, they provide valuable insights into our past, help us navigate the present, and guide us toward a better future. Professionals in these domains engage in a wide range of cognitive activities, including data collection, analysis, interpretation, research, curation, and interdisciplinary collaboration. Their work encompasses diverse cognitive tasks ranging from perception to reasoning to planning. The advent of artificial intelligence (AI) presents new opportunities for these fields, revolutionising the way data is processed, analysed, and interpreted.
Classical AI focuses on representing knowledge using symbols and rules and employs logic and inference to manipulate these symbols and make decisions. In contrast, machine learning-based AI learns patterns and makes predictions based on data. While the rule-based AI offers control, interpretability, and customisation based on human expertise, but may lack scalability and adaptability, the machine learning-based AI provides scalability, adaptability, and the ability to capture complex patterns from data, but can be less transparent and requires substantial training data. Discriminative AI and Generative AI are two primary types of machine learning-based AI. Discriminative AI focuses on learning the decision boundary between different classes in the data, enabling tasks like detection, classification, or prediction. A few examples of this are image classification, sentiment analysis, speech recognition, anomaly detection, etc. In contrast, Generative AI models aim to learn the underlying data distribution and generate new instances that resemble the original data. They can create new content or data points by sampling from the learned distribution. Text generation, image generation, music compositions, video synthesis, etc., are some examples.
As the fields of cultural studies, history, archaeology, art history, and museology embraced digitisation and adopted several computing technologies, AI emerged as a powerful ally, transforming the way data is processed, analysed, and interpreted. AI applications, such as computer vision, text mining, pattern recognition, and natural language processing, enable more efficient and insightful exploration of archaeological sites, artistic styles, historical contexts, and museum collections. Furthermore, applications like chatbots, robot guides, language translation, virtual artwork, art restoration, VR experiences enhance user experiences, provide personalised interactions, and generate new content.
Computer Vision can detect patterns or anomalies that could hint at hidden archaeological sites. This has been demonstrated in numerous significant discoveries across various cultures and regions, including those in Egyptian, Mayan, European, Cambodian, and Indian archaeology. Museums like The Metropolitan Museum of Art in New York are using computer vision to automatically generate tags for images to increase the searchability of their digitised images. It is revolutionising artifact identification and classification processes. It can help to identify, classify, and even authenticate artifacts based on their visual characteristics such as shape, colour, and texture. A notable example is the automatic classification of Mayan pottery based on its stylistic elements. Artificial Neural Networks have been used to categorise art images by movement or style, to classify paintings by artist or era, and to extract and analyse artists’ brushstrokes. Researchers at Rutgers University used deep learning algorithms to analyse artistic styles and techniques throughout history.
Computer Vision techniques are used in the digital restoration of damaged structures, paintings, artifacts, or photographs. By identifying damaged areas and reconstructing the missing parts based on the existing ones, researchers can visualise the original form of the artifact or structure, as was done in the digital restoration of the Renaissance paintings in the Uffizi Gallery in Italy, and frescoes in the Ajanta Caves in India. The restoration scientists at Rijksmuseum in Amsterdam tapped AI to restore the missing elements to digitally reconstruct an ultra-high-resolution image of “The Night Watch” by Rembrandt. The AI solution, ‘Ithaca’, a deep neural network, can restore the missing text of damaged inscriptions and assist historians with identifying the original location and date of creation of such inscriptions.
By analysing multiple images from different angles, computer vision algorithms can reconstruct a 3D model of the site or artifact, such as the 3D mapping of the Great Wall of China. Furthermore, in late 2021, the EU funded an ambitious project, RePAIR (Reconstructing the Past: Artificial Intelligence and Robotics meet Cultural Heritage) which aims to develop intelligent robotic systems to autonomously process, match and physically assemble large fractured using AI-powered image recognition.
AI-powered text mining and natural language processing (NLP) techniques have opened new avenues for analysing vast collections of historical texts. For example, researchers use text mining to analyse historic newspapers archived by the Chronicling America project to identify trends and patterns in news reporting, advertising, and public opinion during different periods. Text mining is employed to study the linguistic features of historical documents to reveal interesting insights into language evolution. It enables researchers to analyse the literature and explore patterns in style, themes, and influences across various time periods and cultural contexts. For example, the Culturomics project at Harvard University discovers cultural trends using text mining on a massive corpus of digitised books. Text mining of historical literature and other textual sources has been widely used to provide valuable insights on various issues such as the portrayal of marginalised groups. It has been applied to analyse artistic and architectural descriptions, historical speeches and debates, religious texts and beliefs, historical medical texts, and propaganda and media coverage.
NLP has been applied to aid in the preservation of endangered languages. Machine learning algorithms are now capable of translating texts between languages. For instance, the Rosetta and Perseus projects incorporate NLP techniques for translation, categorisation, language pattern analysis, and text annotation. It has been instrumental in deciphering ancient scripts. For example, the decipherment of the ancient Hittite language was assisted using NLP techniques, which enabled the researchers to identify word patterns and grammatical structures. It has been applied to the analysis of oral histories. For example, NLP has been applied to transcribe, index, and analyse the testimonies archived in the Shoah Foundation’s Visual History Archive, which contains around 55,000 video testimonies of Holocaust survivors.
Museums have embraced AI applications to enhance visitor experiences. AI-driven chatbots are emerging as useful tools for museums, especially after the pandemic. These chatbots can provide context-specific information, answer queries, and even recommend exhibits based on the visitor’s and/or caller’s preferences. Some museums like Cooper Hewitt Museum in New York, the Heinz Nixdorf Museums Forum in Germany, the Heracleum Archaeological Museum in Greece, the Anne Frank Museum in Amsterdam, and many others have already deployed their chatbots. Chatbots benefit museums by enhancing visitor engagement, providing 24/7 availability, improving scalability, reducing staffing costs, supporting multilingual communication, collecting visitor insights, offering virtual guided tours, and enhancing accessibility.
The incorporation of AI-powered robots in museums worldwide has expanded the realm of visitor interaction. These are gradually being used to engage visitors, provide exhibit information, answer queries, and even guide tours. For example, the Smithsonian deployed ‘Pepper’ robots, to interact with visitors. Another recent development is the deployment of ‘Telepresence Robots.’ These robots offer remote tours of museums, enabling individuals who cannot physically visit the location to still experience the exhibits. Many museums like the American Museum of Natural History in New York, the Mob Museum in Las Vegas, and the Canada Science and Technology Museum in Ottawa have installed such robots. A few museums have experimented with computer vision for facial recognition to recognise returning visitors and provide recommendations and personalised experiences based on their past visits.
A few museums are using AI-based analysis of social media and marketing data to optimise their outreach and engagement strategies. AI-powered sentiment analysis can enable the understanding of visitor feedback, driving necessary improvements. AI offers the potential for optimising exhibit layout and managing crowd flow, potentially improving the overall visitor experience. AI-driven tools can also enhance museums’ educational programs, offering interactive learning experiences.
Infusion of AI is reshaping the future including that of cultural studies, history, archaeology, art history, and museology. With cutting-edge AI applications like advanced computer vision, natural language processing, and robotics, the possibilities for exploring archaeological sites, understanding artistic styles, and delving into historical contexts are expanding exponentially. The fusion of AI and museology heralds a future where AI-powered tools will also revolutionise visitor experiences, deliver personalised interactions, and optimise outreach and engagement strategies. AI coupled with other computing technologies has begun to redefine the workflows associated with many activities related to the discovery, preservation, conservation, restoration, education, management, or enrichment of cultural heritage.
Expansive integration of AI in disciplines related to cultural heritage calls for a shift in professional competencies and education. Professionals in these fields must acquire skills in applying AI technologies and tools to their disciplines through interdisciplinary collaboration. Educational institutions should adapt curricula to include AI-infused courses, fostering a multidisciplinary approach. By equipping professionals with these skills, we can empower them to maximise the transformative potential of AI to redefine their own workflows, outreach, and user experiences. On the other hand, computer science and AI students should be exposed to the exciting possibilities in these domains and encouraged to engage in collaborative projects and internships with experts in these fields.
Also see:
- Developments in Cultural Informatics – I: Enhancing Museum and Archaeo-heritage Site Experiences with VR and AR
- Developments in Cultural Informatics – II: Transforming Archaeology and Museums with IoT
- Developments in Cultural Informatics – III: Text Mining & Natural Language Processing for Digital Humanities
- Developments in Cultural Informatics – IV: Harnessing GIS for Deeper Understanding of History, Culture, and Archaeology
- Developments in Cultural Informatics – V: Transforming Engagements with History through Computer Vision, Graphics, and Image Processing
- Developments in Cultural Informatics – VI: Mechatronics and Robotics Ushering a New Era for Archaeology and Museums
- Developments in Cultural Informatics – VIII: Digital Audio and Video Technologies for Intangible Cultural Heritage
- Transforming Education with AI: From ELIZA to ChatGPT and Beyond
- AI beyond Chatbots: Revolutionising Civil Engineering
- The Daughter Rejuvenates the Mother: Electrifying Electrical Engineering with AI
- How Engineering Education can embrace ChatGPT?
- Unleashing the Power of AI in Electronics and Communication Engineering
Engineering & Computing Education: Reflections and Ideation 
Posted on May 26, 2023
0