Delocate: Detection and Localization for Deepfake Videos with Randomly-Located Tampered Traces
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866914789660295168 |
|---|---|
| author | Hu, Juan Liao, Xin Gao, Difei Tsutsui, Satoshi Wang, Qian Qin, Zheng Shou, Mike Zheng |
| author_facet | Hu, Juan Liao, Xin Gao, Difei Tsutsui, Satoshi Wang, Qian Qin, Zheng Shou, Mike Zheng |
| contents | Deepfake videos are becoming increasingly realistic, showing few tampering traces on facial areasthat vary between frames. Consequently, existing Deepfake detection methods struggle to detect unknown domain Deepfake videos while accurately locating the tampered region. To address thislimitation, we propose Delocate, a novel Deepfake detection model that can both recognize andlocalize unknown domain Deepfake videos. Ourmethod consists of two stages named recoveringand localization. In the recovering stage, the modelrandomly masks regions of interest (ROIs) and reconstructs real faces without tampering traces, leading to a relatively good recovery effect for realfaces and a poor recovery effect for fake faces. Inthe localization stage, the output of the recoveryphase and the forgery ground truth mask serve assupervision to guide the forgery localization process. This process strategically emphasizes the recovery phase of fake faces with poor recovery, facilitating the localization of tampered regions. Ourextensive experiments on four widely used benchmark datasets demonstrate that Delocate not onlyexcels in localizing tampered areas but also enhances cross-domain detection performance. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2401_13516 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Delocate: Detection and Localization for Deepfake Videos with Randomly-Located Tampered Traces Hu, Juan Liao, Xin Gao, Difei Tsutsui, Satoshi Wang, Qian Qin, Zheng Shou, Mike Zheng Computer Vision and Pattern Recognition Cryptography and Security Deepfake videos are becoming increasingly realistic, showing few tampering traces on facial areasthat vary between frames. Consequently, existing Deepfake detection methods struggle to detect unknown domain Deepfake videos while accurately locating the tampered region. To address thislimitation, we propose Delocate, a novel Deepfake detection model that can both recognize andlocalize unknown domain Deepfake videos. Ourmethod consists of two stages named recoveringand localization. In the recovering stage, the modelrandomly masks regions of interest (ROIs) and reconstructs real faces without tampering traces, leading to a relatively good recovery effect for realfaces and a poor recovery effect for fake faces. Inthe localization stage, the output of the recoveryphase and the forgery ground truth mask serve assupervision to guide the forgery localization process. This process strategically emphasizes the recovery phase of fake faces with poor recovery, facilitating the localization of tampered regions. Ourextensive experiments on four widely used benchmark datasets demonstrate that Delocate not onlyexcels in localizing tampered areas but also enhances cross-domain detection performance. |
| title | Delocate: Detection and Localization for Deepfake Videos with Randomly-Located Tampered Traces |
| topic | Computer Vision and Pattern Recognition Cryptography and Security |
| url | https://arxiv.org/abs/2401.13516 |