Keynote Speakers

Abstract

In the field of digital library, in order to conveniently annotate or capture multimedia other than text, such as images or 3D graphics, traditional computer vision method usually needs to pre-process these multimedia data. Taking images as an example, we must first do edge detection, then segmentation, and then describe the content of each closed contour. With the above description, we know it is equivalent to making local annotations on images. However, edge detection and image segmentation are both ill-posed engineering problems, so the probability of being able to describe the content of unknown images well is very low. Benefiting from the ImageNet developed by Fei-Fei Li and Jia Deng in 2010, deep learning has opened up a new horizon in the fields of computer vision and digital libraries, because many topics that were previously very important in the digital library field, such as image annotation and image retrieval, have become much simpler. The ill-posed engineering problems faced by the traditional low-level computer vision will be solved by the visual dictionary with 2,000 object categories provided by ImageNet, the reason is that for an unknown image, we only need to look it up in a visual dictionary to know the approximate content.

You Only Look Once (YOLO) is a one-stage object detector that can quickly detect the type of objects to be detected from unknown images. YOLO is the best tool to provide the metadata required by digital libraries, such as image annotation, image retrieval, and object detection. YOLOv4 and YOLOv7 are object detection tools developed in Taiwan. It is my great honor to talk about the development process of YOLOv4 and YOLOv7 in this keynote speech.

Biography

Mark Liao received his Ph.D. from Northwestern University, Evanston, Illinois, in 1990. He joined the Institute of Information Science, Academia Sinica, Taiwan in 1991. He received the Young Investigators' Award from Academia Sinica in 1998; the Distinguished Research Award from the National Science Council in 2003, 2010 and 2013; the Academia Sinica Investigator Award in 2010; the TECO Award from the TECO Foundation in 2016, and the 64th Academic Award from the Ministry of Education in 2020. His professional activities include: Editorial Board Member, IEEE Signal Processing Magazine (2010-13); Associate Editor, IEEE Transactions on Image Processing (2009-13), IEEE Transactions on Information Forensics and Security (2009-12), IEEE Transactions on Multimedia (1998-2001), ACM Computing Surveys (2018-2021). He is now a Senior Associate Editor of ACM Computing Surveys (2021-present). He has been a Fellow of the IEEE since 2013.

Research Areas

Content-based Multimedia Retrieval

Machine Learning

Artificial Intelligence

Multimedia Information Processing

Abstract

There is a rich and growing body of research concerned with the concept of data cultures, spread across a wide range of disciplines but without any evidence of a universally adopted shared definition or understanding. The knowledge and expertise of information studies researchers in such areas as information behaviour, digital literacies, information management, digital sustainability and preservation are essential perspectives to bring to bear to address issues relating to data cultures, and indeed to ensure our essential relevance in data-driven environments. In this presentation I will explain the concept of data culture and highlight areas needing further research to fully define and examine its dimensions, with the aim of promoting increased understanding and engagement with data cultures from the information studies community.

Biography

Gillian Oliver is Professor of Information Management in the Department of Human Centred Computing, Faculty of Information Technology, co-Director of the Information Empowered Communities Lab at Monash University, Australia and currently Chair Elect of iSchools.org.  After a successful professional practice career she graduated from Monash University in 2006 with a PhD investigating cross cultural information challenges. Her disciplinary background is archival science, with a specific focus on the management of records (information and data) in current environments.The dominant theme running through her research and teaching relates to cultural influences and the need to represent cultural diversity in the systems and technologies that are developed to create and capture data and information. Her research is strongly interdisciplinary  as she seeks to contribute and apply this cultural dimension to  computer science contexts, resulting in strong collaboration with colleagues in software engineering and cybersecurity.  She is currently leading a Australian $2 million dollar research project investigating the meaning of digital citizenship for marginalised communities in Bangladesh, and exploring initiatives relating to Indigenous data sovereignty in Australia and New Zealand. 

Research Areas

Information and data cultures

Recordkeeping research and education

Digital sustainability

Abstract

Artificial Intelligence is increasingly a strategic priority for governments, research funders and higher education institutions. It also has potential to impact nearly every area of library activity, to some degree. It can be used to enhance access to collections. In the form of chatbots, it can provide front facing services. AI, including data literacy, also demands to be incorporated into the Information and academic literacy programme. AI tools can be used in everyday library work. In this context, libraries need to develop a strategic response. The paper develops a SWOT analysis for libraries' in relation to AI. It then identifies the key questions which libraries need to answer to determine how they prioritise their use of AI.

Biography

Dr Andrew Cox is a senior lecturer at the Information School, University of Sheffield, United Kingdom. He is also currently the convenor of IFLA's AI Special Interest Group. His main research area has been around the response of the information professions to contemporary societal challenges such as new technologies, increasing managerialism, datafication, changing conceptualisations of learning and a perceived social crisis around well-being. He wrote a report for the UK professional body, CILIP, on The impact of AI, machine learning, automation and robotics on the information professions: https://www.cilip.org.uk/general/custom.asp?page=researchreport. His other publications relating to AI can be found @ https://www.sheffield.ac.uk/is/people/academic/andrew-cox

Research Areas

My main research area has been the response of information professions to contemporary societal challenges such as new technologies, increasing managerialism, datafication, changing conceptualisations of learning and a perceived crisis of well-being.

Abstract

30 years have passed since the program announcement of the Digital Library Initiative (DLI) by the National Science Foundation in 1993. DLI strongly affected not only the information technology communities but also libraries, museums, and archives, which are so called memory institutions, e.g., launches of the major digital library conferences such as JCDL, TPDL and ICADL, developments of metadata standards oriented to digital resources, and a movement toward information schools in the LIS school community.

Since then, memory institutions have seen dramatic changes in both information technologies and collections. Memory institutions have established functions and services to build digital collections, mostly in the cultural and historical sectors, and to keep them alive for the future, which is referred to as Digital Archiving in this talk. Initially, digital archiving at memory institutions began with the digitization of their institutional holdings. In response to changes in social, cultural, and technological environments, such as the growth of digital humanities and digital shifts among creators, publishers, and audiences, the domains of digital archiving have been expanded to new domains such as intangible cultural heritage, performing arts, popular culture, social and historical events, and so forth. In parallel, metadata standards and technologies developed for the Internet have enabled memory institutions to build new services for the users, e.g., Europeana. This trend may be seen as a paradigm shift from the services based on the institutional collections to the services based on linking of diverse resources in the networked information environments.

The speaker will first present basic features of digital archiving mainly from the viewpoint of metadata followed by some discussions based on his experiences in his research on metadata for digital archiving of intangible cultural heritage, popular culture and disasters. Then, he will discuss conceptual models of digital archiving designed to cover both tangible and intangible entities followed by some open questions.

Biography

Shigeo Sugimoto is a professor emeritus at the University of Tsukuba, Japan. He is an executive director of the Japan Library Association since June 2023. He received his PhD from the Department of Information Science, Faculty of Engineering, Kyoto University, in 1985. He served as a faculty member at the Faculty of Library, Information and Media Science, University of Tsukuba (formerly University of Library and Information Science) for over 30 years. He has been actively working in the research communities of digital libraries and metadata. He has been an active participant of the Dublin Core Metadata Initiative (DCMI) since 1990s and he is currently a member of the Governing Board of DCMI. He has been involved in major international conferences such as ICADL, JCDL, TPDL and iPRES for many years. He was a steering committee chair of ICADL from 2013 to 2018. He has been involved in the international information school community since early 2000. Besides, he was the chair of the Asia Pacific chapter of iSchools from July 2016 to March 2018.

Research Areas

Technological foundations and data models for metadata, digital archiving and digital preservation