Show simple item record

dc.contributor.authorOffert, Fabian
dc.contributor.authorBell, Peter
dc.contributor.editorReussner, Ralf H.
dc.contributor.editorKoziolek, Anne
dc.contributor.editorHeinrich, Robert
dc.date.accessioned2021-01-27T13:33:21Z
dc.date.available2021-01-27T13:33:21Z
dc.date.issued2021
dc.identifier.isbn978-3-88579-701-2
dc.identifier.issn1617-5468
dc.identifier.urihttp://dl.gi.de/handle/20.500.12116/34711
dc.description.abstractMachine vision systems based on deep convolutional neural networks are increasingly utilized in digital humanities projects, particularly in the context of art-historical and audiovisual data. As research has shown, such systems are highly susceptible to bias. We propose that this is not only due to their reliance on biased datasets but also because their perceptual topology, their specific way of representing the visual world, gives rise to a new class of bias that we call perceptual bias. Perceptual bias, we argue, affects almost all currently available “off-the-shelf” machine vision systems, and is thus especially relevant for digital humanities applications, which often rely on such systems for hypothesis building. We evaluate the nature and scope of perceptual bias by means of a close reading of a visual analytics technique called “feature visualization” and propose to understand the development of critical visual analytics techniques as an important (digital) humanities challenge, situated at the interface of computer science and visual studies.en
dc.language.isoen
dc.publisherGesellschaft für Informatik, Bonn
dc.relation.ispartofINFORMATIK 2020
dc.relation.ispartofseriesLecture Notes in Informatics (LNI) - Proceedings, Volume P-307
dc.subjectmachine learning
dc.subjectvisual analytics
dc.subjectcomputer vision
dc.subjectbias
dc.subjectinterpretability
dc.subjectdigital art history
dc.titleUnderstanding Perceptual Bias in Machine Vision Systemsen
mci.reference.pages1295-1305
mci.conference.sessiontitleMethoden und Anwendungen der Computational Humanities
mci.conference.locationKarlsruhe
mci.conference.date28. September - 2. Oktober 2020
dc.identifier.doi10.18420/inf2020_121


Files in this item

Thumbnail

Show simple item record