Recent advances in deep generative models have enabled the generation of realistic synthetic media or deepfakes, including synthetic images, videos, and text. However, synthetic media can be misused for malicious purposes and damage users' trust in online content. This dissertation aims to address several key challenges in defending against the misuse of synthetic media. Key contributions of this dissertation include the following: (1) Understanding challenges with the real-world applicability of existing synthetic media defenses. We curate synthetic videos and text from the wild, i.e., the Internet community, and assess the effectiveness of state-of-the-art defenses on synthetic content in the wild. In addition, we propose practical low-cost adversarial attacks, and systematically measure the adversarial robustness of existing defenses. Our findings reveal that most defenses show significant degradation in performance under real-world detection scenarios, which leads to the second thread of my work: (2) Building detection schemes with improved generalization performance and robustness for synthetic content. Most existing synthetic image detection schemes are highly content-specific, e.g., designed for only human faces, thus limiting their applicability. I propose an unsupervised content-agnostic detection scheme called NoiseScope, which does not require a priori access to synthetic images and is applicable to a wide variety of generative models, i.e., GANs. NoiseScope is also resilient against a range of countermeasures conducted by a knowledgeable attacker. For the text modality, our study reveals that state-of-the-art defenses that mine sequential patterns in the text using Transformer models are vulnerable to simple evasion schemes. We conduct further exploration towards enhancing the robustness of synthetic text detection by leveraging semantic features. / Doctor of Philosophy / Recent advances in deep generative models have enabled the generation of realistic synthetic media or deepfakes, including synthetic images, videos, and text. However, synthetic media can be misused for malicious purposes and damage users' trust in online content. This dissertation aims to address several key challenges in defending against the misuse of synthetic media. Key contributions of this dissertation include the following: (1) Understanding challenges with the real-world applicability of existing synthetic media defenses. We curate synthetic videos and text from the Internet community, and assess the effectiveness of state-of-the-art defenses on the collected datasets. In addition, we systematically measure the robustness of existing defenses by designing practical low-cost attacks, such as changing the configuration of generative models. Our findings reveal that most defenses show significant degradation in performance under real-world detection scenarios, which leads to the second thread of my work: (2) Building detection schemes with improved generalization performance and robustness for synthetic content. Many existing synthetic image detection schemes make decisions by looking for anomalous patterns in a specific type of high-level content, e.g., human faces, thus limiting their applicability. I propose a blind content-agnostic detection scheme called NoiseScope, which does not require synthetic images for training, and is applicable to a wide variety of generative models. For the text modality, our study reveals that state-of-the-art defenses that mine sequential patterns in the text using Transformer models are not robust against simple attacks. We conduct further exploration towards enhancing the robustness of synthetic text detection by leveraging semantic features.

Recent advancements in generative models have resulted in the improvement of hyper- realistic synthetic images or "deepfakes" at high resolutions, making them almost indistin- guishable from real images from cameras. While exciting, this technology introduces room for abuse. Deepfakes have already been misused to produce pornography, political propaganda, and misinformation. The ability to produce fully synthetic content that can cause such mis- information demands for robust deepfake detection frameworks. Most deepfake detection methods are trained in a supervised manner, and fail to generalize to deepfakes produced by newer and superior generative models. More importantly, such detection methods are usually focused on detecting deepfakes having a specific type of content, e.g., face deepfakes. How- ever, other types of deepfakes are starting to emerge, e.g., deepfakes of biomedical images, satellite imagery, people, and objects shown in different settings. Taking these challenges into account, we propose NoiseLearner, an unsupervised and content-agnostic deepfake im- age detection method. NoiseLearner aims to detect any deepfake image regardless of the generative model of origin or the content of the image. We perform a comprehensive evalu- ation by testing on multiple deepfake datasets composed of different generative models and different content groups, such as faces, satellite images, landscapes, and animals. Further- more, we include more recent state-of-the-art generative models in our evaluation, such as StyleGAN3 and probabilistic denoising diffusion models (DDPM). We observe that Noise- Learner performs well on multiple datasets, achieving 96% accuracy on both StyleGAN and StyleGAN2 datasets. / Master of Science / Images synthesized by artificial intelligence, commonly known as deepfakes, are starting to become indistinguishable from real images. While these technological advances are exciting with regards to what a computer can do, it is important to understand that such technol- ogy is currently being used with ill intent. Thus, identifying these images is becoming a growing necessity, especially as deepfake technology grows to perfectly mimic the nature of real images. Current deepfake detection approaches fail to detect deepfakes of other content, such as sattelite imagery or X-rays, and cannot generalize to deepfakes synthesized by new artificial intelligence. Taking these concerns into account, we propose NoiseLearner, a deep- fake detection method that can detect any deepfake regardless of the content and artificial intelligence model used to synthesize it. The key idea behind NoiseLearner is that it does not require any deepfakes to train. Instead, NoiseLearner learns the key features of real images and uses them to differentiate between deepfakes and real images – without ever looking at a single deepfake. Even with this strong constraint, NoiseLearner shows promise by detecting deepfakes of diverse contents and models used to generate them. We also explore different ways to improve NoiseLearner.

<p dir="ltr">The rapid development of machine learning techniques makes it possible to manipulate or synthesize video and audio information while introducing nearly indetectable artifacts. Most media forensics methods analyze the high-level data (e.g., pixels from videos, temporal signals from audios) decoded from compressed media data. Since media manipulation or synthesis methods usually aim to improve the quality of such high-level data directly, acquiring forensic evidence from these data has become increasingly challenging. In this work, we focus on media forensics techniques using the metadata in media formats, which includes container metadata and coding parameters in the encoded bitstream. Since many media manipulation and synthesis methods do not attempt to hide metadata traces, it is possible to use them for forensics tasks. First, we present a video forensics technique using metadata embedded in MP4/MOV video containers. Our proposed method achieved high performance in video manipulation detection, source device attribution, social media attribution, and manipulation tool identification on publicly available datasets. Second, we present a transformer neural network based MP3 audio forensics technique using low-level codec information. Our proposed method can localize multiple compressed segments in MP3 files. The localization accuracy of our proposed method is higher compared to other methods. Third, we present an H.264-based video device matching method. This method can determine if the two video sequences are captured by the same device even if the method has never encountered the device. Our proposed method achieved good performance in a three-fold cross validation scheme on a publicly available video forensics dataset containing 35 devices. Fourth, we present a Graph Neural Network (GNN) based approach for the analysis of MP4/MOV metadata trees. The proposed method is trained using Self-Supervised Learning (SSL), which increased the robustness of the proposed method and makes it capable of handling missing/unseen data. Fifth, we present an efficient approach to compute the spectrogram feature with MP3 compressed audio signals. The proposed approach decreases the complexity of speech feature computation by ~77.6% and saves ~37.87% of MP3 decoding time. The resulting spectrogram features lead to higher synthetic speech detection performance.</p>