| Title |
Deepfake Detection Using Machine Learning Techniques |
| Abstract |
The proliferation of deepfake technology has raised significant concerns regarding its potential misuse in society. This review paper provides a comprehensive overview of the advancements in deepfake detection using machine learning techniques. It analyzes various methodologies, datasets, challenges, and future directions in this field.
CNNs, which are well-known for their efficiency in extracting spatial characteristics, and GNNs, which are adept at acquiring relational information, together constitute a substantial breakthrough in the creation of increasingly complex and dependable deepfake detection methods. By merging the spatial and relational signals included in multimedia content, these hybrid models show enhanced discriminative performance and offer a comprehensive understanding of the intricate changes present in deepfake content.
By carefully reviewing the corpus of previous research, this study summarizes the various benefits of hybrid models and elucidates their potential for addressing the intricate issues raised by synthetic media manipulation. Interestingly, these models' resilience to adversarial assaults is strengthened by their ability to detect minute inconsistencies created by deepfake operations thanks to the mixing of geographical and relational data
|
| Keywords |
deepfake, machine learning, face detection, video editing, deep neural network, deep learning, speech recognition, faceforensics, machine learning |
| Reserch Area |
Engineering |
| Reserch Paper |
Final AIJFR2301005 - V1 I1 - 36-43.pdf |
| Author(s) |
Asmitha Shukla |
| Country |
India |
