• Anubhaw Mishra
• Rohit Sawant
• Prajwal Mohite
• Nilam Thorat

With the latest progress in generative adversarial networks, it's now possible to make incredibly realistic deepfakes. This has raised concerns about misinformation and trust issues. The traditional methods for detecting manipulation struggle to keep up with these advanced deepfake techniques. So, we need strong deep learning solutions. However, a big challenge is creating models that can effectively spot deepfakes from various sources. In this paper, we suggest a new type of deep convolutional neural network (D-CNN) that's designed to learn visual features that can reliably detect deepfakes made using different methods.The D-CNN is trained on a wide range of deepfake types and real examples. Its goal is to accurately identify both familiar and unfamiliar deepfake techniques. This is achieved by using binary cross-entropy loss and the Adam optimizer to improve its learning abilities. The study emphasizes the importance of being able to detect deepfakes in general, which helps protect the integrity of digital media and reduces the spread of misleading synthetic content.

Deep Convolutional Neural Network (D-CNN), Deepfake, Detect

"GUARDING AUTHENTICITY: DETECTION FOR DEEPFAKE", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 5, page no.n543-n550, May-2024, Available :http://www.jetir.org/papers/JETIR2405D71.pdf

"GUARDING AUTHENTICITY: DETECTION FOR DEEPFAKE", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 5, page no. ppn543-n550, May-2024, Available at : http://www.jetir.org/papers/JETIR2405D71.pdf