Combating Digitally Altered Images: Deepfakes Detection
Deepfakes are hyper-realistic but fabricated images, videos or audio files generated using artificial intelligence. In recent times, they have become increasingly accessible and sophisticated. This ease of creation has led to their misuse in spreading misinformation, committing identity theft, and violating personal privacy.
Recognizing the urgent need to address these threats, Saksham Kumar and Rhythm Narang both Microsoft Learn Student Ambassadors, built a Deepfake Detection solution as part of the program’s Quarterly AI Project. This project provides a robust solution for detecting and classifying deepfake content.
Leveraging the power of Azure Machine Learning (Azure ML), this state-of-the-art model can accurately identify deepfakes, thereby helping to mitigate the risks associated with AI-driven misinformation. Love to try it out? Check it out – https://github.com/Polymath-Saksh/deepfake
Additional Resources
Join the Microsoft Learn Student Ambassadors program – https://studentambassadors.com/
Build Generative AI Applications with .NET - https://aka.ms/genai.net
Build Generative AI Applications – https://aka.ms/buildgenai
Get Started Building AI Agents – https://aka.ms/buildaiagents
00:00 Introduction
01:13 Overview of the AI project, objectives and goals
03:35 Architecture and design of the AI project
06:21 Step-by-step walkthrough of the AI solution
09:13 Next steps and future enhancements
11:30 Lessons learnt and best practices
12:30 Resources
13:55 Final remarks for anyone looking to try this out
[eventID:25654]