Scribd
Upload
10 views

Deepfake Detection System Using AI

The document outlines a project titled 'Deep Fake Detection System Using AI' developed by a team from Avanthi Institute of Engineering and Technology. It proposes an automated system utilizing Convolutional Neural Networks (CNNs) to effectively detect deepfakes, overcoming the limitations of traditional manual detection methods. The system aims to provide high accuracy and speed for real-time applications in media verification and digital forensics.

Uploaded by

dineshjillala
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
10 views

Deepfake Detection System Using AI

The document outlines a project titled 'Deep Fake Detection System Using AI' developed by a team from Avanthi Institute of Engineering and Technology. It proposes an automated system utilizing Convolutional Neural Networks (CNNs) to effectively detect deepfakes, overcoming the limitations of traditional manual detection methods. The system aims to provide high accuracy and speed for real-time applications in media verification and digital forensics.

Uploaded by

dineshjillala
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 11

AVANTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY

(Approved by AICTE,New Delhi,Recg. By Govt. of T.S& Affiliated to JNTUH, Hyderabad) NAAC “A” Accredited Institute
Gunthapally(V), Abdullapurmet (M), RR Dist, Near Ramoji Film City, Hyderabad -501512. www.aietg.ac.in Email: [email protected]

Real - Time Research Project


TITLE: DEEP FAKE DETECTION SYSTEM USING AI

TEAM MEMBERS: SANDENENI NANDINI-23Q61A6213


JILLALA DINESH-23Q61A6209
ANUGU RAGHAVENDRA-23Q61A6201
BAKKAYOLLA SAMPATH-23Q61A6203
BARKAM ANVESH- 23Q61A6205

GUIDE NAME: Dr. N. V. RAMANA REDDY


ASSOCIATE PROFESSOR
ABSTRACT

As Deepfakes are AI-generated synthetic media that closely resemble real individuals, posing threats like
misinformation, fraud, and identity theft.

Traditional manual detection methods are slow, outdated, and unable to handle the volume and sophistication of modern
deepfakes.

This project proposes an automated deepfake detection system using Convolutional Neural Networks (CNNs).

The system extracts video frames, detects facial regions, and analyzes them for digital
artifacts and inconsistencies invisible to the human eye.

With high accuracy and speed, the AI-based system distinguishes between real and fake videos.
EXISTING SYSTEM
• Most raditional deepfake detection relies heavily on manual forensic analysis by video and media experts.
• Manual review is time-consuming, labor-intensive, and impractical for real-time applications or large-scale content
platforms.
• Some tools use rule-based algorithms to detect simple visual artifacts such as:
• Irregular eye blinking or facial movements
• Misaligned facial features or distorted lips

DRAWBACKS OF EXISTING SYSTEM


• Manual deepfake detection is time-consuming and inefficient.
• Existing methods are not scalable for handling large volumes of media content.
• Most systems lack the capability for real-time detection and response.
PROPOSED SYSTEM
• The system uses Artificial Intelligence and Deep Learning to detect manipulated video content (deepfakes).
• A Convolutional Neural Network (CNN) is trained on large datasets of real and fake videos to learn identifying features.
• The process begins with video upload by the user through a simple interface.
• The video is broken down into frames, and faces are detected in each frame.
• Each detected face is preprocessed (resized, normalized) and passed to the trained AI model.
• The CNN analyzes facial features for signs of tampering such as:

ADVANTAGES OF PROPOSING SYSTEM


 Provides high accuracy in detecting deepfakes using advanced deep learning models.
 Fully automated detection process, reducing the need for manual analysis.
HARDWARE SYSTEM CONFIGURATION

❖ Processor: Intel Core i7 or higher

❖ RAM: Minimum 16 GB

❖ Hard Disk: At least 100 GB of free storage

❖ GPU: NVIDIA GeForce RTX 3050 or equivalent with 4GB VRAM (for model training and inference)

SOFTWARE REQUIREMENTS

❖ Programming Language: Python

❖ Deep Learning Libraries: TensorFlow, Keras, OpenCV, NumPy

❖ Web Framework: Flask or Django (for building the user interface)

❖ Client-Side Technologies: HTML, CSS, JavaScript


SYSTEM ARCHITECTURE
DATA FLOW DIAGRAM LEVEL 0
DATA FLOW DIAGRAM LEVEL 1
USE CASE DIAGRAM
SEQUENCE DIAGRAM
Conclusion
The DeepFake Detection System using Artificial Intelligence provides a powerful and reliable
solution to combat the growing threat of deepfakes in digital media. By leveraging deep learning
techniques such as Convolutional Neural Networks, the system can effectively identify subtle visual
inconsistencies that are often missed by the human eye. It offers high accuracy, speed, and
adaptability, making it suitable for real-time applications in areas such as media verification, digital
forensics, and online content moderation. This project not only contributes to enhancing content
authenticity but also helps in building trust and security

Thank.Q

You might also like