ANIMAFEST PRO | ANIMAFEST SCANNER XI | Panel 4: Authorship In/Of AI Animation

AniVision: Using AI for Studying Animation Stylistics - Monica Apellaniz Portos, Claudius Stemmler (Germany)

PANEL 4 - AUTHORSHIP IN/OF AI ANIMATION
05/06 SRI 15:20-15:50 KIC

Animators work in many different contexts. In some, they are the key contributors, credited in ways that allow them to claim authorship of their work. In others, their work receives little or no recognition. The latter category includes so-called ephemeral films, which are nonfiction films produced for a specific, usually short-term purpose, such as educational films, commercials, or public service announcements (Prelinger 2006, Reichert 2009, Cook & Thompson 2019). After serving their purpose, these films often end up in archives, barely noticed by scholars and forgotten by the public. Attempts to remedy this problem have always struggled with both the sheer volume of ephemeral films produced and their relative lack of documentation. This problem is further exacerbated when researching something even more specific, such as animation. As a result, apart from a few high-profile individuals, animators working in ephemeral film have remained comparatively anonymous, unable to claim authorship of their work. In recent decades, however, technological advances have suggested the possibility of improving this situation. First, archives have digitized more of their film holdings, making them more accessible. Second, there have been major developments in the field of artificial intelligence, with tools developed through machine learning showing great proficiency in evaluating data for specific tasks. Encouraged by these developments, animation scholars and computer scientists from the St. Pölten University of Applied Sciences (Austria) and the University of Tübingen (Germany) joined forces in a research project called AniVision. Guided by the motto ‘We make animation visible!’, the digital humanities project AniVision uses machine learning and computer vision to explore a large corpus of ephemeral films produced in Austria, East Germany, and West Germany during the Cold War period. An interactive, computer-assisted approach will be used to investigate stylistic patterns of animation within this corpus. Together with an accompanying database, this will allow us to shed light on forgotten animators and write a more comprehensive history of animation in these countries.

Monica Apellaniz Portos is a Junior Researcher in the Media Computing Research group at the University of Applied Sciences of St. Pölten, Austria. She holds a master's degree in AI from the Polytechnic University of Madrid, Spain, and has a rich experience in computer vision applied to diverse domains. Currently, she focuses on visual retrieval, and interactive and explainable machine learning.

Claudius Stemmler obtained a B.A. in film studies from the Johannes Gutenberg University of Mainz. Subsequently, he graduated in media studies at Siegen University. Since then, he has worked on his dissertation about Japanese video game designer Hideo Kojima, a project initially funded by a scholarship from Siegen University’s House of Young Talents Postgraduate Centre. Since February 2023, he has been working as an academic researcher in the Institute of Media Studies’ AniVision project.