Deepfake swapped face detection based on double attention
In view of the existing Deepfake detection algorithms, such problems as low accuracy and poor interpretability are common.A neural network model combining the double attention was proposed, which used channel attention to capture the abnormal features of false faces and combined the location of spat...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
POSTS&TELECOM PRESS Co., LTD
2021-04-01
|
| Series: | 网络与信息安全学报 |
| Subjects: | |
| Online Access: | http://www.cjnis.com.cn/thesisDetails#10.11959/j.issn.2096-109x.2021032 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841529868154568704 |
|---|---|
| author | Xiaojuan GONG Tianqiang HUANG Bin WENG Feng YE Chao XU Lijun YOU |
| author_facet | Xiaojuan GONG Tianqiang HUANG Bin WENG Feng YE Chao XU Lijun YOU |
| author_sort | Xiaojuan GONG |
| collection | DOAJ |
| description | In view of the existing Deepfake detection algorithms, such problems as low accuracy and poor interpretability are common.A neural network model combining the double attention was proposed, which used channel attention to capture the abnormal features of false faces and combined the location of spatial attention to focus the abnormal features.To fully learn the contextual semantic information of the abnormal part of the false face, so as to improve the effectiveness and accuracy of face changing detection.In addition, the decision-making area of real and fake faces was shown effectively in the form of thermal diagram, which provided a certain degree of explanation for the face exchange detection model.Experiments on FaceForensics ++ open source data set show that the detection accuracy of proposed method is superior to MesoInception, Capsule-Forensics and XceptionNet. |
| format | Article |
| id | doaj-art-37d5df9f205e40f5962f801f7560af9f |
| institution | Kabale University |
| issn | 2096-109X |
| language | English |
| publishDate | 2021-04-01 |
| publisher | POSTS&TELECOM PRESS Co., LTD |
| record_format | Article |
| series | 网络与信息安全学报 |
| spelling | doaj-art-37d5df9f205e40f5962f801f7560af9f2025-01-15T03:15:01ZengPOSTS&TELECOM PRESS Co., LTD网络与信息安全学报2096-109X2021-04-01715116059566416Deepfake swapped face detection based on double attentionXiaojuan GONGTianqiang HUANGBin WENGFeng YEChao XULijun YOUIn view of the existing Deepfake detection algorithms, such problems as low accuracy and poor interpretability are common.A neural network model combining the double attention was proposed, which used channel attention to capture the abnormal features of false faces and combined the location of spatial attention to focus the abnormal features.To fully learn the contextual semantic information of the abnormal part of the false face, so as to improve the effectiveness and accuracy of face changing detection.In addition, the decision-making area of real and fake faces was shown effectively in the form of thermal diagram, which provided a certain degree of explanation for the face exchange detection model.Experiments on FaceForensics ++ open source data set show that the detection accuracy of proposed method is superior to MesoInception, Capsule-Forensics and XceptionNet.http://www.cjnis.com.cn/thesisDetails#10.11959/j.issn.2096-109x.2021032Deepfakeface swap detectionfake face detectionattention |
| spellingShingle | Xiaojuan GONG Tianqiang HUANG Bin WENG Feng YE Chao XU Lijun YOU Deepfake swapped face detection based on double attention 网络与信息安全学报 Deepfake face swap detection fake face detection attention |
| title | Deepfake swapped face detection based on double attention |
| title_full | Deepfake swapped face detection based on double attention |
| title_fullStr | Deepfake swapped face detection based on double attention |
| title_full_unstemmed | Deepfake swapped face detection based on double attention |
| title_short | Deepfake swapped face detection based on double attention |
| title_sort | deepfake swapped face detection based on double attention |
| topic | Deepfake face swap detection fake face detection attention |
| url | http://www.cjnis.com.cn/thesisDetails#10.11959/j.issn.2096-109x.2021032 |
| work_keys_str_mv | AT xiaojuangong deepfakeswappedfacedetectionbasedondoubleattention AT tianqianghuang deepfakeswappedfacedetectionbasedondoubleattention AT binweng deepfakeswappedfacedetectionbasedondoubleattention AT fengye deepfakeswappedfacedetectionbasedondoubleattention AT chaoxu deepfakeswappedfacedetectionbasedondoubleattention AT lijunyou deepfakeswappedfacedetectionbasedondoubleattention |