Vid-Freeze: Protecting Images from Malicious Image-to-Video Generation via Temporal Freezing
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_ | 1866918426916683776 |
|---|---|
| author | Chowdhury, Rohit Bala, Aniruddha Jaiswal, Rohan Roheda, Siddharth |
| author_facet | Chowdhury, Rohit Bala, Aniruddha Jaiswal, Rohan Roheda, Siddharth |
| contents | The rapid progress of image-to-video (I2V) generation models has introduced significant risks by enabling deceptive or malicious video synthesis from a single image. Prior defenses such as I2VGuard attempt to immunize images by inducing spatio-temporal degradation, which does not necessarily provide meaningful protection, since residual motion can still convey malicious intent. In this work, we introduce Vid-Freeze -- a novel adversarial defense that adds imperceptible perturbations to enforce temporal freezing in generated videos. Our method explicitly targets attention dynamics in I2V models to suppress motion synthesis. As a result, immunized images produce standstill or near-static videos, effectively blocking malicious content generation. Experiments demonstrate strong protection across models and support temporal freezing as a promising direction for proactive and meaningful defense against I2V misuse. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_23279 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Vid-Freeze: Protecting Images from Malicious Image-to-Video Generation via Temporal Freezing Chowdhury, Rohit Bala, Aniruddha Jaiswal, Rohan Roheda, Siddharth Computer Vision and Pattern Recognition Artificial Intelligence The rapid progress of image-to-video (I2V) generation models has introduced significant risks by enabling deceptive or malicious video synthesis from a single image. Prior defenses such as I2VGuard attempt to immunize images by inducing spatio-temporal degradation, which does not necessarily provide meaningful protection, since residual motion can still convey malicious intent. In this work, we introduce Vid-Freeze -- a novel adversarial defense that adds imperceptible perturbations to enforce temporal freezing in generated videos. Our method explicitly targets attention dynamics in I2V models to suppress motion synthesis. As a result, immunized images produce standstill or near-static videos, effectively blocking malicious content generation. Experiments demonstrate strong protection across models and support temporal freezing as a promising direction for proactive and meaningful defense against I2V misuse. |
| title | Vid-Freeze: Protecting Images from Malicious Image-to-Video Generation via Temporal Freezing |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2509.23279 |