%0 Conference Proceedings
%T Automatic Privacy Classification of Personal Photos
%+ Ludwig Maximilian University [Munich] = Ludwig Maximilians Universität München (LMU)
%+ Deutsches Forschungszentrum für Künstliche Intelligenz GmbH = German Research Center for Artificial Intelligence (DFKI)
%A Buschek, Daniel
%A Bader, Moritz
%A Zezschwitz, Emanuel, Von
%A Luca, Alexander, De
%< avec comité de lecture
%( Lecture Notes in Computer Science
%B 15th Human-Computer Interaction (INTERACT)
%C Bamberg, Germany
%3 Human-Computer Interaction – INTERACT 2015
%V LNCS-9297
%N Part II
%P 428-435
%8 2015-09-14
%D 2015
%R 10.1007/978-3-319-22668-2_33
%K Photos
%K Privacy
%K Classification
%K Images
%K Metadata
%Z Computer Science [cs]Conference papers
%X Tagging photos with privacy-related labels, such as “myself”, “friends” or “public”, allows users to selectively display pictures appropriate in the current situation (e.g. on the bus) or for specific groups (e.g. in a social network). However, manual labelling is time-consuming or not feasible for large collections. Therefore, we present an approach to automatically assign photos to privacy classes. We further demonstrate a study method to gather relevant image data without violating participants’ privacy. In a field study with 16 participants, each user assigned 150 personal photos to self-defined privacy classes. Based on this data, we show that a machine learning approach extracting easily available metadata and visual features can assign photos to user-defined privacy classes with a mean accuracy of 79.38 %.
%G English
%Z TC 13
%2 https://inria.hal.science/hal-01599867/document
%2 https://inria.hal.science/hal-01599867/file/346942_1_En_33_Chapter.pdf
%L hal-01599867
%U https://inria.hal.science/hal-01599867
%~ IFIP-LNCS
%~ IFIP
%~ IFIP-TC13
%~ IFIP-INTERACT
%~ IFIP-LNCS-9297