Wednesday, February 22, 2023, 10:00

STGS seminar room (Werner-von-Siemens-Str. 61, room 02.21)

Register here to attend via Zoom

When science meets art: How computer technologies impact art history

Recent years have seen an increasing intersection between science and art. Computational methods have been developed and employed to study art images, visualize large image corpora, or critically reevaluate terminology and methods. In this talk, Dr. Lang will present previous and current research in the field of digital art history and shows how computer technologies are impacting art history. It highlights the potential of computer scientists and art historians working together but also challenges which such interdisciplinary endeavors face.

Past research projects were driven by questions such as how computational methods can assist art historians and what new insights into terminology or methods they can generate. Art historians study artistic relationships and reception processes to understand the dissemination of specific motifs or styles throughout time and space. In order to reconstruct (cultural) processes on a large scale, researchers utilize models for image and object detection. In this talk, the focus will be on the development of an interactive user-interface which was built to assist with the detection of identical and similar images and objects. Neural networks modeled after the human brain have been used for the task of style transfer, i.e., when a real image is changed to match the style of an artist. Style transfer has provided computer vision with powerful models; in turn, art history has benefited by gaining new insights into artistic style: What is style? Which features define a style? And how does style impact the effect of an artwork? Dr. Lang will also reflect on current research on gaps in art history and in particular provenance research. Are current data models equipped to record such gaps or must they be adapted? Gaps caused by legal restrictions or technical limitations are considered, as well as structure-related gaps, e.g., gender data gaps.

Sabine Lang is a post-doctoral researcher in FAU’s Department Digital Humanities and Social Studies. Her research interests lie in modern art and digital provenance research, with a special focus on data representation and the questions how computer-based methods can be utilized for image analysis and how digital methods impact research and our society as a whole.

Image credit:

1. Kotovenko, A. Sanakoyeu et al. “A Content Transformation Block for Image Style Transfer.” Computer Vision and Pattern Recognition (2019)

Περισσότερα εδώ.