Deepfake Face
Deepfake faces, realistic artificial images or videos of people's faces, are a growing concern due to their potential for misuse in misinformation and identity theft. Current research focuses on developing robust detection methods, employing convolutional neural networks (CNNs) like XceptionNet and EfficientNet, and exploring multi-modal approaches that analyze audio and visual cues simultaneously. This work is crucial for safeguarding digital media authenticity and combating the spread of manipulated content, with ongoing efforts to improve detection accuracy and generalizability across various deepfake generation techniques.
Papers
Towards Identity-Aware Cross-Modal Retrieval: a Dataset and a Baseline
Nicola Messina, Lucia Vadicamo, Leo Maltese, Claudio Gennaro
Generalize Your Face Forgery Detectors: An Insertable Adaptation Module Is All You Need
Xiaotian Si, Linghui Li, Liwei Zhang, Ziduo Guo, Kaiguo Yuan, Bingyu Li, Xiaoyong Li