Deep-Fake-Detection

• Dataset:

  • The dataset can be found and downloaded from: https://www.kaggle.com/competitions/deepfake-detection-challenge/data
  • The code expects it to be put under a folder called /data

• How to run:

  • Download the data and placed it under a /data folder
  • Do "pip install -r reqs.txt" to download the needed packages
  • Run the notebook

• Project rules:

  • The project accounts the 40 % of your grade (working code + written report + presentation).
  • You will get a grade based on the novelty, quality and complexity of the project, along with the ability to show use of appropriate models, preprocessing techniques and evaluation strategies.
  • The projects are done in groups of 2-4.

• These are the list of projects suggested but you are free to choose your own topic as well:

  • Deep fake detection challenge
    • We will use the dataset from https://www.kaggle.com/competitions/deepfake-detection-challenge/overview to an external site.
    • Explore different deep neural network architectures
    • Copying existing notebooks will result in disqualification
    • The solution should be using tensorflow APIs or pytorch or any other framework