• Aishwarya Sedamkar
• Poornima Mishra
• Anuradha Pandey

Detecting deepfake images and videos requires a mix of sophisticated techniques to spot inconsistencies and artifacts that come from AI-generated content. Pixel-level analysis focuses on identifying unnatural textures and lighting, while facial landmark detection looks for misalignments in facial features. Deep learning models, especially convolutional neural networks (CNNs), are designed to recognize subtle differences between real and synthetic media. In videos, temporal inconsistencies like mismatched lip-syncing or unusual facial movements can indicate manipulation. Furthermore, examining the frequency domain for strange noise patterns and checking metadata and compression artifacts can also aid in uncovering deep fakes. Behavioral and contextual analysis seeks out unnatural speech patterns or movements, while cross-modal verification ensures that visual and audio cues are in sync. Tools such as FaceForensics++ and XceptionNet, along with multimodal learning, play a role in identifying these manipulations. However, as deep face technology evolves, detection methods need to adapt to remain effective, presenting an ongoing challenge in combating synthetic media

"AI DEEPFAKE IMAGE & VIDEO DETECTION USING PYTHON", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 2, page no.e798-e800, February-2025, Available :http://www.jetir.org/papers/JETIR2502495.pdf

"AI DEEPFAKE IMAGE & VIDEO DETECTION USING PYTHON", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 2, page no. ppe798-e800, February-2025, Available at : http://www.jetir.org/papers/JETIR2502495.pdf