Compression-Aware Hybrid Framework for Deep Fake Detection in Low-Quality Video

Deep fakes pose a growing threat to digital media integrity by generating highly realistic fake videos that are difficult to detect, especially under the high compression levels commonly used on social media platforms. These compression artifacts often degrade the performance of deep fake detectors,...

Full description

Saved in:
Bibliographic Details
Main Authors: Lagsoun Abdel Motalib, Oujaoura Mustapha, Hedabou Mustapha
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11095666/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849419841935507456
author Lagsoun Abdel Motalib
Oujaoura Mustapha
Hedabou Mustapha
author_facet Lagsoun Abdel Motalib
Oujaoura Mustapha
Hedabou Mustapha
author_sort Lagsoun Abdel Motalib
collection DOAJ
description Deep fakes pose a growing threat to digital media integrity by generating highly realistic fake videos that are difficult to detect, especially under the high compression levels commonly used on social media platforms. These compression artifacts often degrade the performance of deep fake detectors, making reliable detection even more challenging. In this paper, we propose a handcrafted deep fake detection framework that integrates wavelet transforms and Conv3D-based spatiotemporal descriptors for feature extraction, followed by a lightweight ResNet-inspired classifier. Unlike end-to-end deep neural networks, our method emphasizes interpretability and computational efficiency, while maintaining high detection accuracy under diverse real-world conditions. We evaluated four configurations based on input modality and attention mechanisms: RGB with attention, RGB without attention, grayscale with attention, and grayscale without attention. Experiments were conducted on the FaceForensics++ dataset (C23 and C40 compression levels) and Celeb-DF v2 (C0 and C40), across intra- and inter-compression settings, as well as cross-dataset scenarios. Results show that RGB inputs without attention achieve the highest accuracy on FaceForensics++, while grayscale inputs without attention perform best in cross-dataset evaluations on Celeb-DF v2, attaining strong AUC scores. Despite its handcrafted nature, our approach matches or surpasses the existing state-of-the-art (SOTA) methods. Grad-CAM visualizations further reveal both strengths and failures (e.g., occlusion and misalignment), offering valuable insights for refinement. These findings underscore the potential of our framework for efficient and effective deep fake detection in low-resource and real-time environments.
format Article
id doaj-art-b0c122d2d4e844cc80f4fc8306c6e400
institution Kabale University
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-b0c122d2d4e844cc80f4fc8306c6e4002025-08-20T03:31:56ZengIEEEIEEE Access2169-35362025-01-011313198013199710.1109/ACCESS.2025.359235811095666Compression-Aware Hybrid Framework for Deep Fake Detection in Low-Quality VideoLagsoun Abdel Motalib0https://orcid.org/0009-0006-9196-1166Oujaoura Mustapha1Hedabou Mustapha2https://orcid.org/0000-0002-7872-4469Department of Computer Sciences, Networks, and Telecoms (GIRT), Laboratory of Mathematics, Informatics, and Communication Systems (MISCOM), National School of Applied Sciences (ENSA), Cadi Ayyad University, Safi, MoroccoDepartment of Computer Sciences, Networks, and Telecoms (GIRT), Laboratory of Mathematics, Informatics, and Communication Systems (MISCOM), National School of Applied Sciences (ENSA), Cadi Ayyad University, Safi, MoroccoCollege of Computing, Mohammed VI Polytechnic University (UM6P), Ben Guerir, MoroccoDeep fakes pose a growing threat to digital media integrity by generating highly realistic fake videos that are difficult to detect, especially under the high compression levels commonly used on social media platforms. These compression artifacts often degrade the performance of deep fake detectors, making reliable detection even more challenging. In this paper, we propose a handcrafted deep fake detection framework that integrates wavelet transforms and Conv3D-based spatiotemporal descriptors for feature extraction, followed by a lightweight ResNet-inspired classifier. Unlike end-to-end deep neural networks, our method emphasizes interpretability and computational efficiency, while maintaining high detection accuracy under diverse real-world conditions. We evaluated four configurations based on input modality and attention mechanisms: RGB with attention, RGB without attention, grayscale with attention, and grayscale without attention. Experiments were conducted on the FaceForensics++ dataset (C23 and C40 compression levels) and Celeb-DF v2 (C0 and C40), across intra- and inter-compression settings, as well as cross-dataset scenarios. Results show that RGB inputs without attention achieve the highest accuracy on FaceForensics++, while grayscale inputs without attention perform best in cross-dataset evaluations on Celeb-DF v2, attaining strong AUC scores. Despite its handcrafted nature, our approach matches or surpasses the existing state-of-the-art (SOTA) methods. Grad-CAM visualizations further reveal both strengths and failures (e.g., occlusion and misalignment), offering valuable insights for refinement. These findings underscore the potential of our framework for efficient and effective deep fake detection in low-resource and real-time environments.https://ieeexplore.ieee.org/document/11095666/Deep fake detectionhandcrafted featureswavelet transformResNet-based classifiercompression robustness
spellingShingle Lagsoun Abdel Motalib
Oujaoura Mustapha
Hedabou Mustapha
Compression-Aware Hybrid Framework for Deep Fake Detection in Low-Quality Video
IEEE Access
Deep fake detection
handcrafted features
wavelet transform
ResNet-based classifier
compression robustness
title Compression-Aware Hybrid Framework for Deep Fake Detection in Low-Quality Video
title_full Compression-Aware Hybrid Framework for Deep Fake Detection in Low-Quality Video
title_fullStr Compression-Aware Hybrid Framework for Deep Fake Detection in Low-Quality Video
title_full_unstemmed Compression-Aware Hybrid Framework for Deep Fake Detection in Low-Quality Video
title_short Compression-Aware Hybrid Framework for Deep Fake Detection in Low-Quality Video
title_sort compression aware hybrid framework for deep fake detection in low quality video
topic Deep fake detection
handcrafted features
wavelet transform
ResNet-based classifier
compression robustness
url https://ieeexplore.ieee.org/document/11095666/
work_keys_str_mv AT lagsounabdelmotalib compressionawarehybridframeworkfordeepfakedetectioninlowqualityvideo
AT oujaouramustapha compressionawarehybridframeworkfordeepfakedetectioninlowqualityvideo
AT hedaboumustapha compressionawarehybridframeworkfordeepfakedetectioninlowqualityvideo