DFFMD: A Deepfake Face Mask Dataset for Infectious Disease Era With Deepfake Detection Algorithms

Supervisor:

Asst.Prof. JITHIKA M

Team Members

Muhammed sadhef PM(STM20CS039)
Sahila noora ibrahim(STM20CS049)
Neha fazal CK(STM20CS042)
Shibiliya KK(STM20CS028)

Description

Deepfake Face Mask Dataset (DFFMD) is based on a novel
Inception-ResNet-v2 model with preprocessing stages, feature-based
approaches, residual connections, and batch normalization techniques.
Deepfake detection models extract intricate features from both spatial and
temporal dimensions of videos.
These features are then compared to established norms and patterns in
genuine content.
The system employs deep neural networks to autonomously learn
distinctive features that differentiate between genuine and manipulated
content.