Enregistré dans:
| Auteurs principaux: | , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | https://arxiv.org/abs/2403.12710 |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866929282217934848 |
|---|---|
| author | Ilic, Filip Zhao, He Pock, Thomas Wildes, Richard P. |
| author_facet | Ilic, Filip Zhao, He Pock, Thomas Wildes, Richard P. |
| contents | Concerns for the privacy of individuals captured in public imagery have led to privacy-preserving action recognition. Existing approaches often suffer from issues arising through obfuscation being applied globally and a lack of interpretability. Global obfuscation hides privacy sensitive regions, but also contextual regions important for action recognition. Lack of interpretability erodes trust in these new technologies. We highlight the limitations of current paradigms and propose a solution: Human selected privacy templates that yield interpretability by design, an obfuscation scheme that selectively hides attributes and also induces temporal consistency, which is important in action recognition. Our approach is architecture agnostic and directly modifies input imagery, while existing approaches generally require architecture training. Our approach offers more flexibility, as no retraining is required, and outperforms alternatives on three widely used datasets. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2403_12710 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Selective, Interpretable, and Motion Consistent Privacy Attribute Obfuscation for Action Recognition Ilic, Filip Zhao, He Pock, Thomas Wildes, Richard P. Computer Vision and Pattern Recognition Machine Learning Concerns for the privacy of individuals captured in public imagery have led to privacy-preserving action recognition. Existing approaches often suffer from issues arising through obfuscation being applied globally and a lack of interpretability. Global obfuscation hides privacy sensitive regions, but also contextual regions important for action recognition. Lack of interpretability erodes trust in these new technologies. We highlight the limitations of current paradigms and propose a solution: Human selected privacy templates that yield interpretability by design, an obfuscation scheme that selectively hides attributes and also induces temporal consistency, which is important in action recognition. Our approach is architecture agnostic and directly modifies input imagery, while existing approaches generally require architecture training. Our approach offers more flexibility, as no retraining is required, and outperforms alternatives on three widely used datasets. |
| title | Selective, Interpretable, and Motion Consistent Privacy Attribute Obfuscation for Action Recognition |
| topic | Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2403.12710 |