EfficientNet with transformer integrations for enhancing deepfake detections
The emergence of Artificial Intelligence has brought about significant advancements, yet it also raises concerns, notably regarding the proliferation of "deepfakes" – digitally altered facial media that blurs the line between reality and fabrication. Extensive research has been conducted t...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project (FYP) |
| Language: | English |
| Published: |
Nanyang Technological University
2024
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/175419 |
| _version_ | 1826119295035244544 |
|---|---|
| author | Varsha, Saravanabavan |
| author2 | Deepu Rajan |
| author_facet | Deepu Rajan Varsha, Saravanabavan |
| author_sort | Varsha, Saravanabavan |
| collection | NTU |
| description | The emergence of Artificial Intelligence has brought about significant advancements, yet it also raises concerns, notably regarding the proliferation of "deepfakes" – digitally altered facial media that blurs the line between reality and fabrication. Extensive research has been conducted to develop methods for detecting deepfakes, aiming to mitigate their adverse impacts.
This study investigates popular deepfake generation and detection techniques, assessing their effectiveness and limitations. Building upon this, the architectures and performances of EfficientNetB4 were explored and integrated with Spatial and Vision Transformer to create two resource-efficient and accurate deepfake detection models trained on the FaceForensics++ dataset. Evaluation metrics such as accuracy, F1 score, and Area Under Curve (AUC) were employed to compare the proposed architectures with existing models. The proposed EfficientNetB4 models integrated with Spatial Transformer and Vision Transformer have achieved accuracies of 87.50% and 91.43% respectively. |
| first_indexed | 2024-10-01T04:57:33Z |
| format | Final Year Project (FYP) |
| id | ntu-10356/175419 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T04:57:33Z |
| publishDate | 2024 |
| publisher | Nanyang Technological University |
| record_format | dspace |
| spelling | ntu-10356/1754192024-05-10T15:39:52Z EfficientNet with transformer integrations for enhancing deepfake detections Varsha, Saravanabavan Deepu Rajan School of Computer Science and Engineering ASDRajan@ntu.edu.sg Computer and Information Science Deepfakes Efficientnet Transformer The emergence of Artificial Intelligence has brought about significant advancements, yet it also raises concerns, notably regarding the proliferation of "deepfakes" – digitally altered facial media that blurs the line between reality and fabrication. Extensive research has been conducted to develop methods for detecting deepfakes, aiming to mitigate their adverse impacts. This study investigates popular deepfake generation and detection techniques, assessing their effectiveness and limitations. Building upon this, the architectures and performances of EfficientNetB4 were explored and integrated with Spatial and Vision Transformer to create two resource-efficient and accurate deepfake detection models trained on the FaceForensics++ dataset. Evaluation metrics such as accuracy, F1 score, and Area Under Curve (AUC) were employed to compare the proposed architectures with existing models. The proposed EfficientNetB4 models integrated with Spatial Transformer and Vision Transformer have achieved accuracies of 87.50% and 91.43% respectively. Bachelor's degree 2024-04-24T02:06:47Z 2024-04-24T02:06:47Z 2024 Final Year Project (FYP) Varsha, S. (2024). EfficientNet with transformer integrations for enhancing deepfake detections. Final Year Project (FYP), Nanyang Technological University, Singapore. https://hdl.handle.net/10356/175419 https://hdl.handle.net/10356/175419 en SCSE23-0535 application/pdf Nanyang Technological University |
| spellingShingle | Computer and Information Science Deepfakes Efficientnet Transformer Varsha, Saravanabavan EfficientNet with transformer integrations for enhancing deepfake detections |
| title | EfficientNet with transformer integrations for enhancing deepfake detections |
| title_full | EfficientNet with transformer integrations for enhancing deepfake detections |
| title_fullStr | EfficientNet with transformer integrations for enhancing deepfake detections |
| title_full_unstemmed | EfficientNet with transformer integrations for enhancing deepfake detections |
| title_short | EfficientNet with transformer integrations for enhancing deepfake detections |
| title_sort | efficientnet with transformer integrations for enhancing deepfake detections |
| topic | Computer and Information Science Deepfakes Efficientnet Transformer |
| url | https://hdl.handle.net/10356/175419 |
| work_keys_str_mv | AT varshasaravanabavan efficientnetwithtransformerintegrationsforenhancingdeepfakedetections |