Zero-shot image privacy classification with Vision-Language Models
Fuente:
arXiv
Guardado en:
| Autores principales: | , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866908586260561920 |
|---|---|
| author | Baia, Alina Elena Xompero, Alessio Cavallaro, Andrea |
| author_facet | Baia, Alina Elena Xompero, Alessio Cavallaro, Andrea |
| contents | While specialized learning-based models have historically dominated image privacy prediction, the current literature increasingly favours adopting large Vision-Language Models (VLMs) designed for generic tasks. This trend risks overlooking the performance ceiling set by purpose-built models due to a lack of systematic evaluation. To address this problem, we establish a zero-shot benchmark for image privacy classification, enabling a fair comparison. We evaluate the top-3 open-source VLMs, according to a privacy benchmark, using task-aligned prompts and we contrast their performance, efficiency, and robustness against established vision-only and multi-modal methods. Counter-intuitively, our results show that VLMs, despite their resource-intensive nature in terms of high parameter count and slower inference, currently lag behind specialized, smaller models in privacy prediction accuracy. We also find that VLMs exhibit higher robustness to image perturbations. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_09253 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Zero-shot image privacy classification with Vision-Language Models Baia, Alina Elena Xompero, Alessio Cavallaro, Andrea Computer Vision and Pattern Recognition Machine Learning Multimedia While specialized learning-based models have historically dominated image privacy prediction, the current literature increasingly favours adopting large Vision-Language Models (VLMs) designed for generic tasks. This trend risks overlooking the performance ceiling set by purpose-built models due to a lack of systematic evaluation. To address this problem, we establish a zero-shot benchmark for image privacy classification, enabling a fair comparison. We evaluate the top-3 open-source VLMs, according to a privacy benchmark, using task-aligned prompts and we contrast their performance, efficiency, and robustness against established vision-only and multi-modal methods. Counter-intuitively, our results show that VLMs, despite their resource-intensive nature in terms of high parameter count and slower inference, currently lag behind specialized, smaller models in privacy prediction accuracy. We also find that VLMs exhibit higher robustness to image perturbations. |
| title | Zero-shot image privacy classification with Vision-Language Models |
| topic | Computer Vision and Pattern Recognition Machine Learning Multimedia |
| url | https://arxiv.org/abs/2510.09253 |