Thanks to visit codestin.com
Credit goes to www.scribd.com

Scribd
Upload
5 views1 page

ABSTRACT

The document discusses the rise of deepfakes due to advancements in deep learning, highlighting concerns over identity theft, misinformation, and public trust. It emphasizes the need for technological, legislative, and awareness measures to combat these issues and presents a prototype that uses various deep learning architectures for detecting manipulated media. VGG19 and MobileNet are identified as effective models for image and video prediction, respectively.

Uploaded by

O593 DEEPIKA
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
5 views1 page

ABSTRACT

The document discusses the rise of deepfakes due to advancements in deep learning, highlighting concerns over identity theft, misinformation, and public trust. It emphasizes the need for technological, legislative, and awareness measures to combat these issues and presents a prototype that uses various deep learning architectures for detecting manipulated media. VGG19 and MobileNet are identified as effective models for image and video prediction, respectively.

Uploaded by

O593 DEEPIKA
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 1

ABSTRACT

The rapid advancement of deep learning has led to the creation of realistic
deepfakes, sparking concerns about identity theft, scams, privacy breaches, and the
spread of misinformation, impacting public trust and psychological well-being.
Addressing these challenges necessitates a blend of technological innovations,
legislative actions, and public awareness initiatives. DeepFakes, generated using
algorithms like Generative Adversarial Networks (GAN) and Autoencoders. The
fabricated images are so lifelike that detecting them with the naked eye is nearly
impossible. Deepfakes can be used to spread misinformation, manipulate public
opinion, damage reputations, and even create fraudulent content. Deepfakes blur
the lines between reality and fiction,making it increasingly challenging to discern
truth from falsity, which can undermine trust in media and institutions. Therefore,
it's crucial to recognize and construct a model capable of accurately differentiating
between genuine and altered media. The prototype explores deep learning's
efficacy in detecting manipulated images and videos, utilizing advanced
architectures like VGG19, MobileNet, ResNet, MesoNet, CNN, and Xception.
VGG19 is chosen for image prediction with the Celeb-df dataset due to its superior
accuracy over other models. Meanwhile, MobileNet is selected for video prediction
with the DFDC dataset for its accuracy and efficiency compared to alternatives
like VGG19 and MesoNet.

You might also like