BIV-Priv-Seg: Locating Private Content in Images Taken by People With Visual Impairments
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866916560473423872 |
|---|---|
| author | Tseng, Yu-Yun Sharma, Tanusree Zhang, Lotus Stangl, Abigale Findlater, Leah Wang, Yang Gurari, Danna |
| author_facet | Tseng, Yu-Yun Sharma, Tanusree Zhang, Lotus Stangl, Abigale Findlater, Leah Wang, Yang Gurari, Danna |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_18243 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | BIV-Priv-Seg: Locating Private Content in Images Taken by People With Visual Impairments Tseng, Yu-Yun Sharma, Tanusree Zhang, Lotus Stangl, Abigale Findlater, Leah Wang, Yang Gurari, Danna Computer Vision and Pattern Recognition 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. |
| title | BIV-Priv-Seg: Locating Private Content in Images Taken by People With Visual Impairments |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2407.18243 |