Automatic Privacy Classification of Personal Photos
Automatic Privacy Classification of Personal Photos
复制标题
DOI：
10.1007/978-3-319-22668-2_33
复制
发表时间：
2015-09
期刊：
影响因子：
--
通讯作者：
Daniel Buschek;Moritz Bader;E. V. Zezschwitz;A. D. Luca
中科院分区：
文献类型：
--
作者：
Daniel Buschek;Moritz Bader;E. V. Zezschwitz;A. D. Luca
使用隐私相关标签（例如“我自己”、“朋友”或“公众”）标记照片，允许用户选择性地显示适合当前情况（例如在公共汽车上）或特定群体（例如在社交网络中）的图片。然而，手工标记很耗时，或者对于大量收集不可行。因此，我们提出了一种方法来自动分配照片的隐私类。我们进一步展示了一种在不侵犯参与者隐私的情况下收集相关图像数据的研究方法。在一项有16名参与者的实地研究中，每个用户将150张个人照片分配到自定义的隐私类别中。基于这些数据，我们表明，提取容易获得的元数据和视觉特征的机器学习方法可以将照片分配给用户定义的隐私类，平均准确率为79.38%。
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 %.