Team 28

This software design description describes the architecture and system design of deep fake video detection.Deep fake is a type of sysnethetic media in which a person’s face is replaced with that of someone else in an existing photograph or video. its algorithm uses artificial intelligence and machine learning to modify visual and audio content. Deep fake entails the training of generative neural networks such as “autoencoders” and “GANS” (generative adversarial networks) . Several social media networks have already taken the initative and built technology to detect deep fake content. A component that can be reused a cross multiple system and applications can be packaged and distributed as an API , an API to connect to a particular database. In the API will be contain a combine architecture and used a combine dataset. API allow services and products to communicate with each other

This software design description describes deep fake video detection system design and provides the main design viewpoint of the system to communicate to key design stake holders.The project will attempt to evaluate videos across several platforms in order to avoid the spread of incorrect information.the project will train with combine architecture like ( resnet,siamese,xceptionnet) and combine dataset. the convolution neural network (CNN) technique will be used in this project.

• introduction
• System Overview
• Design Viewpoints
• Data Design
• API Design
• Requirements Matrix

The intended audience of this document are the clients and supervisors of the API system.