Home > Research > Publications & Outputs > PUDD: Towards Robust Multi-modal Prototype-base...

Electronic data

  • PUDD

    Accepted author manuscript, 3.53 MB, PDF document

    Available under license: CC BY: Creative Commons Attribution 4.0 International License

View graph of relations

PUDD: Towards Robust Multi-modal Prototype-based Deepfake Detection

Research output: Contribution to conference - Without ISBN/ISSN Conference paperpeer-review

Forthcoming

Standard

PUDD: Towards Robust Multi-modal Prototype-based Deepfake Detection. / Lopez Pellicer, Alvaro; Li, Yi; Angelov, Plamen.
2024. Paper presented at IEEE / CVF Computer Vision and Pattern Recognition Conference (CVPR) 2024, Seattle, Washington, United States.

Research output: Contribution to conference - Without ISBN/ISSN Conference paperpeer-review

Harvard

Lopez Pellicer, A, Li, Y & Angelov, P 2024, 'PUDD: Towards Robust Multi-modal Prototype-based Deepfake Detection', Paper presented at IEEE / CVF Computer Vision and Pattern Recognition Conference (CVPR) 2024, Seattle, United States, 19/06/24 - 21/06/24.

APA

Lopez Pellicer, A., Li, Y., & Angelov, P. (in press). PUDD: Towards Robust Multi-modal Prototype-based Deepfake Detection. Paper presented at IEEE / CVF Computer Vision and Pattern Recognition Conference (CVPR) 2024, Seattle, Washington, United States.

Vancouver

Lopez Pellicer A, Li Y, Angelov P. PUDD: Towards Robust Multi-modal Prototype-based Deepfake Detection. 2024. Paper presented at IEEE / CVF Computer Vision and Pattern Recognition Conference (CVPR) 2024, Seattle, Washington, United States.

Author

Lopez Pellicer, Alvaro ; Li, Yi ; Angelov, Plamen. / PUDD: Towards Robust Multi-modal Prototype-based Deepfake Detection. Paper presented at IEEE / CVF Computer Vision and Pattern Recognition Conference (CVPR) 2024, Seattle, Washington, United States.9 p.

Bibtex

@conference{6b087366e1004475827522f35c3c2c22,
title = "PUDD: Towards Robust Multi-modal Prototype-based Deepfake Detection",
abstract = "Deepfake techniques generate highly realistic data, mak- ing it challenging for humans to discern between actual and artificially generated images. Recent advancements in deep learning-based deepfake detection methods, particularly with diffusion models, have shown remarkable progress. However, there is a growing demand for real-world appli- cations to detect unseen individuals, deepfake techniques, and scenarios. To address this limitation, we propose a Prototype-based Unified Framework for Deepfake Detec- tion (PUDD). PUDD offers a detection system based on similarity, comparing input data against known prototypes for video classification and identifying potential deepfakes or previously unseen classes by analyzing drops in similar- ity. Our extensive experiments reveal three key findings: (1) PUDD achieves an accuracy of 95.1% on Celeb-DF, out- performing state-of-the-art deepfake detection methods; (2) PUDD leverages image classification as the upstream task during training, demonstrating promising performance in both image classification and deepfake detection tasks dur- ing inference; (3) PUDD requires only 2.7 seconds for re- training on new data and emits 105 times less carbon com- pared to the state-of-the-art model, making it significantly more environmentally friendly.",
author = "{Lopez Pellicer}, Alvaro and Yi Li and Plamen Angelov",
year = "2024",
month = mar,
day = "6",
language = "English",
note = "IEEE / CVF Computer Vision and Pattern Recognition Conference (CVPR) 2024 : 2nd Workshop and Challenge on DeepFake Analysis and Detection, CVPR 2024 ; Conference date: 19-06-2024 Through 21-06-2024",
url = "https://cvpr.thecvf.com",

}

RIS

TY - CONF

T1 - PUDD: Towards Robust Multi-modal Prototype-based Deepfake Detection

AU - Lopez Pellicer, Alvaro

AU - Li, Yi

AU - Angelov, Plamen

PY - 2024/3/6

Y1 - 2024/3/6

N2 - Deepfake techniques generate highly realistic data, mak- ing it challenging for humans to discern between actual and artificially generated images. Recent advancements in deep learning-based deepfake detection methods, particularly with diffusion models, have shown remarkable progress. However, there is a growing demand for real-world appli- cations to detect unseen individuals, deepfake techniques, and scenarios. To address this limitation, we propose a Prototype-based Unified Framework for Deepfake Detec- tion (PUDD). PUDD offers a detection system based on similarity, comparing input data against known prototypes for video classification and identifying potential deepfakes or previously unseen classes by analyzing drops in similar- ity. Our extensive experiments reveal three key findings: (1) PUDD achieves an accuracy of 95.1% on Celeb-DF, out- performing state-of-the-art deepfake detection methods; (2) PUDD leverages image classification as the upstream task during training, demonstrating promising performance in both image classification and deepfake detection tasks dur- ing inference; (3) PUDD requires only 2.7 seconds for re- training on new data and emits 105 times less carbon com- pared to the state-of-the-art model, making it significantly more environmentally friendly.

AB - Deepfake techniques generate highly realistic data, mak- ing it challenging for humans to discern between actual and artificially generated images. Recent advancements in deep learning-based deepfake detection methods, particularly with diffusion models, have shown remarkable progress. However, there is a growing demand for real-world appli- cations to detect unseen individuals, deepfake techniques, and scenarios. To address this limitation, we propose a Prototype-based Unified Framework for Deepfake Detec- tion (PUDD). PUDD offers a detection system based on similarity, comparing input data against known prototypes for video classification and identifying potential deepfakes or previously unseen classes by analyzing drops in similar- ity. Our extensive experiments reveal three key findings: (1) PUDD achieves an accuracy of 95.1% on Celeb-DF, out- performing state-of-the-art deepfake detection methods; (2) PUDD leverages image classification as the upstream task during training, demonstrating promising performance in both image classification and deepfake detection tasks dur- ing inference; (3) PUDD requires only 2.7 seconds for re- training on new data and emits 105 times less carbon com- pared to the state-of-the-art model, making it significantly more environmentally friendly.

M3 - Conference paper

T2 - IEEE / CVF Computer Vision and Pattern Recognition Conference (CVPR) 2024

Y2 - 19 June 2024 through 21 June 2024

ER -