Enhancing MobileNet for efficient deepfake detection: a hybrid approach with GhostNet and ShuffleNet
Deepfake technology, which involves the manipulation of video, audio, or images using artificial intelligence, has become a growing concern due to its potential for misuse in areas such as information security and social manipulation. Despite advances in deepfake detection, existing models are often...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project (FYP) |
| Language: | English |
| Published: |
Nanyang Technological University
2024
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/181143 |
| _version_ | 1826124766960943104 |
|---|---|
| author | Lee, Zheng Xuan |
| author2 | Deepu Rajan |
| author_facet | Deepu Rajan Lee, Zheng Xuan |
| author_sort | Lee, Zheng Xuan |
| collection | NTU |
| description | Deepfake technology, which involves the manipulation of video, audio, or images using artificial intelligence, has become a growing concern due to its potential for misuse in areas such as information security and social manipulation. Despite advances in deepfake detection, existing models are often too computationally expensive to deploy on mobile or resource-constrained devices. This research aims to address this gap by proposing an enhanced MobileNet architecture tailored for efficient deepfake detection. Through the integration of lightweight modules from GhostNet and ShuffleNet and the application of model compression techniques such as pruning and knowledge distillation, this hybrid model achieves high accuracy with significantly reduced computational complexity. Extensive experiments were conducted to evaluate the model across several deepfake datasets and the results demonstrate that the proposed architecture strikes an optimal balance between accuracy, model size and inference time. These findings provide valuable insights for deploying deepfake detection models in resource-constrained environments, including mobile devices, without compromising performance. |
| first_indexed | 2025-03-09T13:51:28Z |
| format | Final Year Project (FYP) |
| id | ntu-10356/181143 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2025-03-09T13:51:28Z |
| publishDate | 2024 |
| publisher | Nanyang Technological University |
| record_format | dspace |
| spelling | ntu-10356/1811432024-11-18T00:17:11Z Enhancing MobileNet for efficient deepfake detection: a hybrid approach with GhostNet and ShuffleNet Lee, Zheng Xuan Deepu Rajan College of Computing and Data Science ASDRajan@ntu.edu.sg Computer and Information Science Deepfake detection Deepfake Mobilenet Ghostnet Shufflenet Knowledge distillation Channel pruning Convolutional neural network Deepfake technology, which involves the manipulation of video, audio, or images using artificial intelligence, has become a growing concern due to its potential for misuse in areas such as information security and social manipulation. Despite advances in deepfake detection, existing models are often too computationally expensive to deploy on mobile or resource-constrained devices. This research aims to address this gap by proposing an enhanced MobileNet architecture tailored for efficient deepfake detection. Through the integration of lightweight modules from GhostNet and ShuffleNet and the application of model compression techniques such as pruning and knowledge distillation, this hybrid model achieves high accuracy with significantly reduced computational complexity. Extensive experiments were conducted to evaluate the model across several deepfake datasets and the results demonstrate that the proposed architecture strikes an optimal balance between accuracy, model size and inference time. These findings provide valuable insights for deploying deepfake detection models in resource-constrained environments, including mobile devices, without compromising performance. Bachelor's degree 2024-11-18T00:17:11Z 2024-11-18T00:17:11Z 2024 Final Year Project (FYP) Lee, Z. X. (2024). Enhancing MobileNet for efficient deepfake detection: a hybrid approach with GhostNet and ShuffleNet. Final Year Project (FYP), Nanyang Technological University, Singapore. https://hdl.handle.net/10356/181143 https://hdl.handle.net/10356/181143 en SCSE23-0857 application/pdf Nanyang Technological University |
| spellingShingle | Computer and Information Science Deepfake detection Deepfake Mobilenet Ghostnet Shufflenet Knowledge distillation Channel pruning Convolutional neural network Lee, Zheng Xuan Enhancing MobileNet for efficient deepfake detection: a hybrid approach with GhostNet and ShuffleNet |
| title | Enhancing MobileNet for efficient deepfake detection: a hybrid approach with GhostNet and ShuffleNet |
| title_full | Enhancing MobileNet for efficient deepfake detection: a hybrid approach with GhostNet and ShuffleNet |
| title_fullStr | Enhancing MobileNet for efficient deepfake detection: a hybrid approach with GhostNet and ShuffleNet |
| title_full_unstemmed | Enhancing MobileNet for efficient deepfake detection: a hybrid approach with GhostNet and ShuffleNet |
| title_short | Enhancing MobileNet for efficient deepfake detection: a hybrid approach with GhostNet and ShuffleNet |
| title_sort | enhancing mobilenet for efficient deepfake detection a hybrid approach with ghostnet and shufflenet |
| topic | Computer and Information Science Deepfake detection Deepfake Mobilenet Ghostnet Shufflenet Knowledge distillation Channel pruning Convolutional neural network |
| url | https://hdl.handle.net/10356/181143 |
| work_keys_str_mv | AT leezhengxuan enhancingmobilenetforefficientdeepfakedetectionahybridapproachwithghostnetandshufflenet |