This is a demo workbook from a MDST legacy project: Real vs Fake face Detection. The workbook was adapted from W24 project session, which was instructed by Sachchit Kunichetty and Zain Zai. If you are interested in this project, feel free to visit here for more resources.
Below is the motivation behind this project:
Deepfakes are everywhere now, and like us you're probably wondering "wow how am I going to figure out if that video of Morgan Freeman telling me he isn't real, is actually real?". While a true deepfake detector is too complicated for a single workbook, we're going to do something a little bit easier - try and determine whether a picture of a face is fake or not. To solve this task, we will be using CNN (Convolutional Neural Networks)
Skills Learned: Machine Learning, Deep Learning, Computer Vision, PyTorch
Naturally, the code for this challenge will be pushed to a different repository on GitHub seperated from your tutorials repository. You will be able to submit both repository link in the forms we send out later.
Get started at RvF.ipynb. DO NOT MODIFY any files in starter_code