Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2405.04133 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911869691756544 |
|---|---|
| author | He, Peisong Zhu, Leyao Li, Jiaxing Wang, Shiqi Li, Haoliang |
| author_facet | He, Peisong Zhu, Leyao Li, Jiaxing Wang, Shiqi Li, Haoliang |
| contents | The generative model has made significant advancements in the creation of realistic videos, which causes security issues. However, this emerging risk has not been adequately addressed due to the absence of a benchmark dataset for AI-generated videos. In this paper, we first construct a video dataset using advanced diffusion-based video generation algorithms with various semantic contents. Besides, typical video lossy operations over network transmission are adopted to generate degraded samples. Then, by analyzing local and global temporal defects of current AI-generated videos, a novel detection framework by adaptively learning local motion information and global appearance variation is constructed to expose fake videos. Finally, experiments are conducted to evaluate the generalization and robustness of different spatial and temporal domain detection methods, where the results can serve as the baseline and demonstrate the research challenge for future studies. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2405_04133 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Exposing AI-generated Videos: A Benchmark Dataset and a Local-and-Global Temporal Defect Based Detection Method He, Peisong Zhu, Leyao Li, Jiaxing Wang, Shiqi Li, Haoliang Computer Vision and Pattern Recognition The generative model has made significant advancements in the creation of realistic videos, which causes security issues. However, this emerging risk has not been adequately addressed due to the absence of a benchmark dataset for AI-generated videos. In this paper, we first construct a video dataset using advanced diffusion-based video generation algorithms with various semantic contents. Besides, typical video lossy operations over network transmission are adopted to generate degraded samples. Then, by analyzing local and global temporal defects of current AI-generated videos, a novel detection framework by adaptively learning local motion information and global appearance variation is constructed to expose fake videos. Finally, experiments are conducted to evaluate the generalization and robustness of different spatial and temporal domain detection methods, where the results can serve as the baseline and demonstrate the research challenge for future studies. |
| title | Exposing AI-generated Videos: A Benchmark Dataset and a Local-and-Global Temporal Defect Based Detection Method |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2405.04133 |