Improving Video Deepfake Detection: A DCT-Based Approach with Patch-Level Analysis

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Guarnera, Luca, Manganello, Salvatore, Battiato, Sebastiano
Format: Preprint
Publié: 2023
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866910290479677440
author Guarnera, Luca
Manganello, Salvatore
Battiato, Sebastiano
author_facet Guarnera, Luca
Manganello, Salvatore
Battiato, Sebastiano
contents A new algorithm for the detection of deepfakes in digital videos is presented. The I-frames were extracted in order to provide faster computation and analysis than approaches described in the literature. To identify the discriminating regions within individual video frames, the entire frame, background, face, eyes, nose, mouth, and face frame were analyzed separately. From the Discrete Cosine Transform (DCT), the Beta components were extracted from the AC coefficients and used as input to standard classifiers. Experimental results show that the eye and mouth regions are those most discriminative and able to determine the nature of the video under analysis.
format Preprint
id arxiv_https___arxiv_org_abs_2310_11204
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Improving Video Deepfake Detection: A DCT-Based Approach with Patch-Level Analysis
Guarnera, Luca
Manganello, Salvatore
Battiato, Sebastiano
Computer Vision and Pattern Recognition
Image and Video Processing
A new algorithm for the detection of deepfakes in digital videos is presented. The I-frames were extracted in order to provide faster computation and analysis than approaches described in the literature. To identify the discriminating regions within individual video frames, the entire frame, background, face, eyes, nose, mouth, and face frame were analyzed separately. From the Discrete Cosine Transform (DCT), the Beta components were extracted from the AC coefficients and used as input to standard classifiers. Experimental results show that the eye and mouth regions are those most discriminative and able to determine the nature of the video under analysis.
title Improving Video Deepfake Detection: A DCT-Based Approach with Patch-Level Analysis
topic Computer Vision and Pattern Recognition
Image and Video Processing
url https://arxiv.org/abs/2310.11204