A multi-color spatio-temporal approach for detecting DeepFake
The current surge in hyper-realistic faces created artificially using DeepFakes necessitates media forensics solutions suited to video streams and perform reliably with a low false alarm rate at the video level. The paper proposes a spatial and temporal aware pipeline to detect DeepFake videos autom...
| Main Authors: | , , , , |
|---|---|
| Format: | Conference or Workshop Item |
| Published: |
2022
|
| Subjects: |
| _version_ | 1796866683578613760 |
|---|---|
| author | Waseem, Saima Abu Bakar, Syed R. Omar, Zaid Ahmed, Bilal Ashfaq Baloch, Saba |
| author_facet | Waseem, Saima Abu Bakar, Syed R. Omar, Zaid Ahmed, Bilal Ashfaq Baloch, Saba |
| author_sort | Waseem, Saima |
| collection | ePrints |
| description | The current surge in hyper-realistic faces created artificially using DeepFakes necessitates media forensics solutions suited to video streams and perform reliably with a low false alarm rate at the video level. The paper proposes a spatial and temporal aware pipeline to detect DeepFake videos automatically. Our method employed a two-stream convolutional neural network to extract local spatial and temporal features independently. These features are then fed to fully connected layers to classify whether a video has been subject to manipulation. The proposed method has been evaluated against FaceForensics++, DFTIMIT, and DFD benchmarks. Our suggested technique demonstrates encouraging performance in this task. |
| first_indexed | 2024-03-05T21:15:51Z |
| format | Conference or Workshop Item |
| id | utm.eprints-98758 |
| institution | Universiti Teknologi Malaysia - ePrints |
| last_indexed | 2024-03-05T21:15:51Z |
| publishDate | 2022 |
| record_format | dspace |
| spelling | utm.eprints-987582023-02-02T08:26:21Z http://eprints.utm.my/98758/ A multi-color spatio-temporal approach for detecting DeepFake Waseem, Saima Abu Bakar, Syed R. Omar, Zaid Ahmed, Bilal Ashfaq Baloch, Saba TK Electrical engineering. Electronics Nuclear engineering The current surge in hyper-realistic faces created artificially using DeepFakes necessitates media forensics solutions suited to video streams and perform reliably with a low false alarm rate at the video level. The paper proposes a spatial and temporal aware pipeline to detect DeepFake videos automatically. Our method employed a two-stream convolutional neural network to extract local spatial and temporal features independently. These features are then fed to fully connected layers to classify whether a video has been subject to manipulation. The proposed method has been evaluated against FaceForensics++, DFTIMIT, and DFD benchmarks. Our suggested technique demonstrates encouraging performance in this task. 2022 Conference or Workshop Item PeerReviewed Waseem, Saima and Abu Bakar, Syed R. and Omar, Zaid and Ahmed, Bilal Ashfaq and Baloch, Saba (2022) A multi-color spatio-temporal approach for detecting DeepFake. In: 12th International Conference on Pattern Recognition Systems, ICPRS 2022, 7 - 10 June 2022, Saint-Etienne, France. http://dx.doi.org/10.1109/ICPRS54038.2022.9853853 |
| spellingShingle | TK Electrical engineering. Electronics Nuclear engineering Waseem, Saima Abu Bakar, Syed R. Omar, Zaid Ahmed, Bilal Ashfaq Baloch, Saba A multi-color spatio-temporal approach for detecting DeepFake |
| title | A multi-color spatio-temporal approach for detecting DeepFake |
| title_full | A multi-color spatio-temporal approach for detecting DeepFake |
| title_fullStr | A multi-color spatio-temporal approach for detecting DeepFake |
| title_full_unstemmed | A multi-color spatio-temporal approach for detecting DeepFake |
| title_short | A multi-color spatio-temporal approach for detecting DeepFake |
| title_sort | multi color spatio temporal approach for detecting deepfake |
| topic | TK Electrical engineering. Electronics Nuclear engineering |
| work_keys_str_mv | AT waseemsaima amulticolorspatiotemporalapproachfordetectingdeepfake AT abubakarsyedr amulticolorspatiotemporalapproachfordetectingdeepfake AT omarzaid amulticolorspatiotemporalapproachfordetectingdeepfake AT ahmedbilalashfaq amulticolorspatiotemporalapproachfordetectingdeepfake AT balochsaba amulticolorspatiotemporalapproachfordetectingdeepfake AT waseemsaima multicolorspatiotemporalapproachfordetectingdeepfake AT abubakarsyedr multicolorspatiotemporalapproachfordetectingdeepfake AT omarzaid multicolorspatiotemporalapproachfordetectingdeepfake AT ahmedbilalashfaq multicolorspatiotemporalapproachfordetectingdeepfake AT balochsaba multicolorspatiotemporalapproachfordetectingdeepfake |