diplomsko delo
Tadej Logar (Author), Peter Peer (Mentor), Borut Batagelj (Co-mentor)

Abstract

V diplomski nalogi se soočamo s problematiko odkrivanja lažnih posnetkov. Lažni posnetki se na spletu pojavljajo vse pogosteje in z uporabo tehnologije globokih ponaredkov (angl. Deepfakes) za ustvarjanje teh posnetkov postajajo tudi tako prepričljivi, da lahko pretentajo ljudi. Cilj globokih ponaredkov je velikokrat širjenje dezinformacij ali omadeževanje ugleda znane osebe. Za namen računalniškega prepoznavanja globokih ponaredkov predstavimo dva sorodna pristopa, ki temeljita na arhitekturi transformerjev in delujeta na osnovi posnetka, za razliko od drugih metod, ki delujejo na osnovi posameznih slik. Imenujeta se Video Vision Transformer (ViViT) in UniFormerV2. Modele teh pristopov smo naučili na podatkovnih zbirkah globokih ponaredkov FaceForensics++ in Celeb-DF-v2. Preizkusili smo tudi zmogljivost modelov na testnem naboru iz zbirke DFDC. S temi modeli smo dosegli rezultate, ki so primerljivi tudi z dosedaj najboljšimi na tem področju. V okviru diplomske naloge opišemo še našo metodologijo, tehnologijo uporabljenih modelov in podrobnosti implementacije. Predstavimo tudi podrobne rezultate, eksperimente ter primerjavo z drugačnimi pristopi pri odkrivanju globokih ponaredkov.

Keywords

odkrivanje globokih ponaredkov;globoki ponaredki;lažni posnetki;Video Vision Transformer;UniFormerV2;univerzitetni študij;diplomske naloge;

Data

Language: Slovenian
Year of publishing:
Typology: 2.11 - Undergraduate Thesis
Organization: UL FRI - Faculty of Computer and Information Science
Publisher: [T. Logar]
UDC: 004.85:7.061(043.2)
COBISS: 190633987 Link will open in a new window
Views: 53
Downloads: 18
Average score: 0 (0 votes)
Metadata: JSON JSON-RDF JSON-LD TURTLE N-TRIPLES XML RDFA MICRODATA DC-XML DC-RDF RDF

Other data

Secondary language: English
Secondary title: Deepfake Detection using Video Transformers
Secondary abstract: In this bachelor's thesis we examine the task of Deepfake detection. These fake videos are appearing online with increasing frequency. With the use of deep learning for their creation, they have become convincing enough to trick humans. The goal of creating these fake videos is often to spread misinformation or damage the reputations of celebrities. For this task of detecting fake videos, we present two related video-based approaches, with each using the transformer architecture. These approaches are known as the Video Vision Transformer (ViViT) and UniFormerV2. We trained models of these two approaches on two datasets of fake videos, FaceForensics++ and Celeb-DF-v2. We also tested the performance of these models on an additional test set of videos from the DFDC dataset. With the use of these models, we have achieved results comparable to state-of-the-art approaches in this field. As part of the thesis, we describe our methodology, the technologies used in the approaches, and certain implementation details. We also present detailed results of the models we trained, our experiments, and a comparison of our results with some of the different approaches to Deepfake detection.
Secondary keywords: deepfake detection;deepfake;deep learning;machine learning;Video Vision Transformer;UniFormerV2;computer science;computer and information science;diploma;Ponarejanje in ponaredki;Globoko učenje (strojno učenje);Računalništvo;Univerzitetna in visokošolska dela;
Type (COBISS): Bachelor thesis/paper
Study programme: 1000468
Embargo end date (OpenAIRE): 1970-01-01
Thesis comment: Univ. v Ljubljani, Fak. za računalništvo in informatiko
Pages: 43 str.
ID: 23187552