A5075315087- Digital Media Forensic Detection
- Generative Adversarial Networks and Image Synthesis
- Anomaly Detection Techniques and Applications
- Network Security and Intrusion Detection
- Image Enhancement Techniques
- Biomedical Text Mining and Ontologies
- Advanced Image Processing Techniques
- Face recognition and analysis
- Adversarial Robustness in Machine Learning
- Topic Modeling
- Advanced Malware Detection Techniques
- Image Retrieval and Classification Techniques
- Internet Traffic Analysis and Secure E-voting
Universitatea Națională de Știință și Tehnologie Politehnica București
2021-2024
Generative models have evolved immensely in the last few years. GAN-based video and image generation has become very accessible due to open source software available anyone, that may pose a threat society. Deepfakes can be used intimidate, blackmail certain public figures or mislead public. At same time, with rising popularity of deepfakes, detection algorithms also significantly. The majority those focus on images rather than explore temporal evolution video. In this paper, we whether...
The advent of generative networks and their adoption in numerous domains communities have led to a wave innovation breakthroughs artificial intelligence machine learning. Generative Adversarial Networks (GANs) expanded the scope what is possible with learning, allowing for new applications areas such as computer vision, natural language processing, creative AI. GANs, particular, been used wide range tasks, including image video generation, data augmentation, style transfer, anomaly...
The crucial effort to counteract deepfakes and misinformation at large holds great importance in our society, especially this moment time. Deepfake detectors evolve the same pace as deepfake generators, or even slower, more than that, they are trained on a limited amount of data do not achieve generalization most situations. primary challenge associated with training lies necessity for substantial number diverse generated samples originating from multitude distinct models — an achievement...
Image generation has seen huge leaps in the last few years. Less than 10 years ago we could not generate accurate images using deep learning at all, and now it is almost impossible for average person to distinguish a real image from generated one. In spite of fact that some amazing use cases, can also be used with ill intent. As an example, deepfakes have become more indistinguishable pictures poses threat society. It important us vigilant active against deepfakes, ensure false information...
Information is everywhere, and sometimes we have no idea if what read, watch or listen accurate, real authentic. This paper focuses on detecting deep learning generated videos, deepfakes - a phenomenon which more present in today's society. While there are some very good methods of deepfakes, two key elements that should always be considered, i.e., method perfect deepfake generation techniques continue to evolve, even faster than detection methods. In our proposed architectures, focus family...
Recent advancements in Generative Adversarial Networks (GANs) have enabled photorealistic image generation with high quality. However, the malicious use of such generated media has raised concerns regarding visual misinformation. Although deepfake detection research demonstrated accuracy, it is vulnerable to advances techniques and adversarial iterations on countermeasures. To address this, we propose a proactive sustainable training augmentation solution that introduces artificial...