|
|
Deepfake Detection in Video Samples
Krumpholc, Jan ; Veigend, Petr (oponent) ; Lapšanský, Tomáš (vedoucí práce)
In last years, we could see increase of internet frauds and forgeries. Starting with easier detectable cases like phishing and fake ads, through social engeneering and disinformation campaigns, and ending with attacks using artificial inteligence: Synthetics media, and especially deepfakes. These attacks are very effective because it's difficult to validate authenticity of deepfake media for basic user, and they are in rise in last few years with availability and effectivity of creation tools for public. This thesis is focused on video deepfakes: What methods are used for their creation, what are their weak points, and mainly, how to find these weaknesses and decide, whenever media is deepfake or not. We will analyze state-of-the-art methods of detecting deepfakes, what are their strengths and weaknesses, and develop possible new methods of detection. In the end we will compare results with modern solutions and evaluate result.
|
|
|
Use of Diffusion Models in Deepfakes
Trúchly, Dominik ; Malinka, Kamil (oponent) ; Lapšanský, Tomáš (vedoucí práce)
A deepfake is a type of synthetic media created through sophisticated machine learning algorithms, particularly deep neural networks. As an example Generative adversarial neural networks (GANs), that are capable of generating images that are almost impossible for ordinary individuals to differentiate from genuine reality. Consequently, deepfake detection algorithms have been developed to address this growing concern. Leveraging advanced machine learning techniques, these algorithms analyze various features within images and videos to identify inconsistencies or anomalies indicative of manipulation. This thesis investigates the application of diffusion models, commonly utilized in digital image processing to enhance image quality by reducing noise and blurring, in bolstering the realism of deepfakes. By using these models, we test their effect on detecting deepfakes images using deepfake detectors.
|
|
|
Diffusion Models and their Impact on Cybersecurity
Dvorščák, Patrik ; Homoliak, Ivan (oponent) ; Lapšanský, Tomáš (vedoucí práce)
This thesis explores the performance of diffusion models (DMs) and generative adversarial networks (GANs) in creating AI-generated visual content across multiple applications, including face synthesis, text-to-image generation, artistic rendering, image-to-image translation, video synthesis, and super-resolution. Through comparative experiments, this research evaluates the models' ability to generate detailed, realistic, and artistically compelling visuals from textual and image prompts. The results reveal that DMs excel in producing highly detailed images that closely follow text prompts, particularly effective in face synthesis and text-to-image tasks. In contrast, GANs are more adept at rendering realistic environmental scenes, suitable for applications requiring immersive visuals. Both model types are competent in artistic rendering, though they differ in style adaptation and creativity. The thesis concludes with future research directions aimed at enhancing model efficacy and integrating these technologies more effectively into practical applications.
|
|
|
Virtuální stroj Petriho sítí
Lapšanský, Tomáš ; Janoušek, Vladimír (oponent) ; Kočí, Radek (vedoucí práce)
Bakalárska práca formálne definuje pojem Objektovo orientované Petriho siete. Práca ďalej navrhuje koncept prekladača a virtuálneho stroja pre Objektovo orientované Petriho siete s využitím jazyk PNTalk. Popisuje implementáciu virtuálneho stroja a prekladača.
|
|
|
Fake Face Detection in the Digital Images
Lapšanský, Tomáš ; Goldmann, Tomáš (oponent) ; Orság, Filip (vedoucí práce)
In recent years, we can observe the rise and rapid development of neural networks and artificial intelligence in information technology, which include deepfake photos and videos. Generative adversarial neural networks (GANs) are a clear example of this. Nowadays, they can achieve virtually impossible results for the average person to distinguish from reality. Since these networks can therefore be misused for various purposes, it is necessary to be able to distinguish between what is generated and what is real. This thesis explores current state-of-the-art neural network solutions that can serve as suitable models for deepfake detection. We investigate individual architectures that are suitable as a baseline model for detection, address possible improvements to this model, and develop several new architectures. We then investigate these and evaluate their results. In conclusion, we have a discussion of the results and open further questions on this complex issue.
|
|
|
Virtuální stroj Petriho sítí
Lapšanský, Tomáš ; Janoušek, Vladimír (oponent) ; Kočí, Radek (vedoucí práce)
Bakalárska práca formálne definuje pojem Objektovo orientované Petriho siete. Práca ďalej navrhuje koncept prekladača a virtuálneho stroja pre Objektovo orientované Petriho siete s využitím jazyk PNTalk. Popisuje implementáciu virtuálneho stroja a prekladača.
|
host ::
RSS.