![References
[1] Guarnera, Luca, Oliver Giudice, and Sebastiano Battiato. "Fighting deepfake by exposing the convolutional
traces on images." IEEE Access 8 (2020): 165085-165098.
[2] Kohli, Aditi, and Abhinav Gupta. "Detecting deepfake, faceswap and face2face facial forgeries using
frequency cnn." Multimedia Tools and Applications 80, no. 12 (2021): 18461-18478.
[3] Ismail, Aya, Marwa Elpeltagy, Mervat S. Zaki, and Kamal Eldahshan. "A new deep learning-based
methodology for video deepfake detection using XGBoost." Sensors 21, no. 16 (2021): 5413.
[4] Taeb, Maryam, and Hongmei Chi. "Comparison of deepfake detection techniques through deep learning."
Journal of Cybersecurity and Privacy 2, no. 1 (2022): 89-106.
[5] Pasupuleti, Venkat Rao, Prasanth Reddy Tathireddy, Gopi Dontagani, and Shaik Abdul Rahim. "Deepfake
Detection Using Custom Densenet." In 2023 14th International Conference on Computing Communication and
Networking Technologies (ICCCNT), pp. 1-5. IEEE, 2023.
[6] Matern, F.; Riess, C.; Stamminger, M. Exploiting visual artifacts to expose deepfakes and face manipulations.
In Proceedings of the 2019 IEEE Winter Applications of Computer Vision Workshops (WACVW), Waikoloa
Village, HI, USA, 1–7 January 2019.
[7] Zhao, Hanqing, Wenbo Zhou, Dongdong Chen, Tianyi Wei, Weiming Zhang, and Nenghai Yu. "Multi-
attentional deepfake detection." In Proceedings of the IEEE/CVF conference on computer vision and pattern
recognition, pp. 2185-2194. 2021.
[8] Pan, Deng, Lixian Sun, Rui Wang, Xingjian Zhang, and Richard O. Sinnott. "Deepfake detection through deep
learning." In 2020 IEEE/ACM International Conference on Big Data Computing, Applications and Technologies
(BDCAT), pp. 134-143. IEEE, 2020.
[9] Nirkin, Yuval, Lior Wolf, Yosi Keller, and Tal Hassner. "Deepfake detection based on the discrepancy between
the face and its context." arXiv preprint arXiv:2008.12262 (2020).
[10] Kohli, Aditi, and Abhinav Gupta. "Detecting deepfake, faceswap and face2face facial forgeries using
frequency cnn." Multimedia Tools and Applications 80, no. 12 (2021): 18461-18478.
8](https://image.slidesharecdn.com/comsnetppt-250127065058-6085e2dd/85/Comsnet-PPt-pptxch-utils-data-import-DataLoader-TensorDatase-8-320.jpg)
Comsnet PPt.pptxch.utils.data import DataLoader, TensorDatase
- 1. COMSNETS 2025 17th International Conference on COMmunication Systems & NETworkS Initiative by COMSNETS Association Bengaluru, India NAKSHATRA Technical Paper/Poster Presentation Event Temporal Deepfake Detection using CNN with Spatio- Temporal Features Soundarya B C Manipal Institute of Technology, Bengaluru Gururaj H L Manipal Institute of Technology, Bengaluru Authors 1
- 2. Introduction ▪ Deepfake uses AI, particularly GANs, to create realistic fake media. ▪ Applications in the entertainment industry, voice restoration, and video creation. ▪ Misinformation, privacy breaches, public safety threats. ▪ Need for reliable detection due to increasing sophistication of deepfake technology. 2
- 3. Research Objectives ▪To develop a robust video-based deepfake detection system. ▪To combine spatial feature extraction with temporal inconsistencies for improved accuracy. ▪To evaluate the performance of the proposed model on standard datasets like DFDC. 3
- 4. Proposed Methodology ▪Input video frames are preprocessed (224x224 resizing, mean subtraction, normalization) ▪Spatial features are extracted using ResNet34 ▪Temporal features are captured using Dense Swin Transformer. 4
- 5. Proposed Methodology ▪The model utilizes Local Ternary Patterns (LTP) to capture texture details by encoding intensity changes as patterns. ▪ Edge detection methods like the Canny edge detector to find unexpected changes in pixel intensity gradients. 5
- 6. Results & Discussion The proposed work uses a machine with a high-end processor (1GHz) and GPU (4GB- NVIDIA GeForce GTX 1050 Ti). Python 3.10 programming language is used in Anaconda Environment. 6
- 7. Conclusion ▪Detection accuracy for other datasets experienced a minor decrease of 2-3% compared to the fine-tuned ResNet34 model ▪Notably, the proposed model demonstrated exceptional accuracy of 98.25% in prediction ▪In future work, we can enhance the evaluation of the models performance by employing cross-validation techniques 7
- 8. References [1] Guarnera, Luca, Oliver Giudice, and Sebastiano Battiato. "Fighting deepfake by exposing the convolutional traces on images." IEEE Access 8 (2020): 165085-165098. [2] Kohli, Aditi, and Abhinav Gupta. "Detecting deepfake, faceswap and face2face facial forgeries using frequency cnn." Multimedia Tools and Applications 80, no. 12 (2021): 18461-18478. [3] Ismail, Aya, Marwa Elpeltagy, Mervat S. Zaki, and Kamal Eldahshan. "A new deep learning-based methodology for video deepfake detection using XGBoost." Sensors 21, no. 16 (2021): 5413. [4] Taeb, Maryam, and Hongmei Chi. "Comparison of deepfake detection techniques through deep learning." Journal of Cybersecurity and Privacy 2, no. 1 (2022): 89-106. [5] Pasupuleti, Venkat Rao, Prasanth Reddy Tathireddy, Gopi Dontagani, and Shaik Abdul Rahim. "Deepfake Detection Using Custom Densenet." In 2023 14th International Conference on Computing Communication and Networking Technologies (ICCCNT), pp. 1-5. IEEE, 2023. [6] Matern, F.; Riess, C.; Stamminger, M. Exploiting visual artifacts to expose deepfakes and face manipulations. In Proceedings of the 2019 IEEE Winter Applications of Computer Vision Workshops (WACVW), Waikoloa Village, HI, USA, 1–7 January 2019. [7] Zhao, Hanqing, Wenbo Zhou, Dongdong Chen, Tianyi Wei, Weiming Zhang, and Nenghai Yu. "Multi- attentional deepfake detection." In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp. 2185-2194. 2021. [8] Pan, Deng, Lixian Sun, Rui Wang, Xingjian Zhang, and Richard O. Sinnott. "Deepfake detection through deep learning." In 2020 IEEE/ACM International Conference on Big Data Computing, Applications and Technologies (BDCAT), pp. 134-143. IEEE, 2020. [9] Nirkin, Yuval, Lior Wolf, Yosi Keller, and Tal Hassner. "Deepfake detection based on the discrepancy between the face and its context." arXiv preprint arXiv:2008.12262 (2020). [10] Kohli, Aditi, and Abhinav Gupta. "Detecting deepfake, faceswap and face2face facial forgeries using frequency cnn." Multimedia Tools and Applications 80, no. 12 (2021): 18461-18478. 8
- 9. 14. ‘Thank You’ Slide (1 slide) Just mention ‘THANK YOU’ on this slide. Include ‘QUESTIONS?’ If you want. THANK YOU Questions? 9