Share these talks and lectures with your colleagues
Invite colleaguesCase study
The TailoredMedia Project: Taming the data beast
Journal of Digital Media Management,
11
(4),
355-367
(2023)
Abstract
The manual tagging of information such as persons, objects or places is a time-consuming task that typically needs trained archivists. This inefficient use of resources leads to disproportionately high resource consumption for a relatively small quantity of tagged material. The TailoredMedia project set out to reverse this problem by using machine learning to tag a high quantity of material and having archivists ensure the quality of the tagged information. This paper describes the integration of a set of analysis tools based on artificial intelligence (AI), along with focused research on specific tools (efficient scene classification, few-shot object detection, scene text detection). The paper also introduces Taylor — an avatar for the system’s AI capabilities, designed as part of the user interface to provide explanations and support the user’s work to validate annotations and search for content.
Keywords: AI-supported AV-mining; user-interfaces; support for research and annotation
The full article is available to subscribers to the journal.
Author's Biography
Christoph Bauer is a workflow developer at ORF — the Austrian Broadcasting Corporation — where he is responsible for preservation, digitisation and restoration within the ORF Multimedia Archives. He is also the Acting System Administrator for Archive Systems and Audiovisual Digitisation. Outside of ORF, he is the current General Secretary of maa (the Media-Archives-Austria Association), and a member of the FIAT/IFTA Digitisation and Migration Commission.
Werner Bailer is a key researcher at DIGITAL — Institute for Digital Technologies at Joanneum Research. His research interests include audiovisual content analysis and retrieval, multimedia metadata and machine learning. He is a member of various working groups for standardisation, and co-chairs the MPEG group on neural network compression. He holds a Dipl.Ing degree in media technology and design, and wrote his diploma thesis on motion estimation and segmentation for film/video standards conversion.
Stefanie Größbacher is a junior researcher in the Institute of Creative Media Technologies’ Media Computing Research Group, based at St. Pölten University of Applied Sciences. Her research focuses on the fields of usability engineering, graphic design, user-centred design and human–computer interaction. She also teaches these areas at bachelor’s level, with a focus on media technology, smart engineering and creative computing.
Peter Judmaier is a professor at the Institute of Creative Media Technologies at St. Pölten University of Applied Sciences. He is a lecturer on both the bachelor’s and master’s study programmes, with a focus on human–computer interaction, augmented/virtual reality and digital game production. He has more than 25 years’ experience in the fields of usability engineering and user experience design, having also worked at University of Vienna, Vienna University of Technology and several IT companies. He earned an interdisciplinary master’s degree in information and communication technology from Vienna University of Technology, and wrote his PhD thesis on e-learning in adult education.
