BIV-Priv-Seg: Locating Private Content in Images Taken by People With Visual Impairments
This addresses privacy risks for people with visual impairments by providing a foundational dataset, though it is incremental as it builds on existing localization methods.
The paper tackles the problem of blind or low vision individuals inadvertently sharing private information in photos by introducing BIV-Priv-Seg, the first dataset with segmentation annotations for 16 private object categories from 1,028 images taken by this group, and finds that modern models struggle with locating non-salient, small, or text-lacking private objects.
Individuals who are blind or have low vision (BLV) are at a heightened risk of sharing private information if they share photographs they have taken. To facilitate developing technologies that can help them preserve privacy, we introduce BIV-Priv-Seg, the first localization dataset originating from people with visual impairments that shows private content. It contains 1,028 images with segmentation annotations for 16 private object categories. We first characterize BIV-Priv-Seg and then evaluate modern models' performance for locating private content in the dataset. We find modern models struggle most with locating private objects that are not salient, small, and lack text as well as recognizing when private content is absent from an image. We facilitate future extensions by sharing our new dataset with the evaluation server at https://vizwiz.org/tasks-and-datasets/object-localization.