Detecting Deepfakes with OpenCV and SVM

prerequisites: intermediate Python • beginner scikit-learn and scikit-image • basics of OpenCV
skills learned: binary classification and evaluation of binary classifiers • SVM classification • facial image processing

Pavel Korshunov

4 weeks · 6-10 hours per week · INTERMEDIATE

filed under

Data Science

This title has been retired and is no longer for sale.

Look inside

In this liveProject, you’ll step into the role of a forensics consultant. You’re investigating a ring of cyber criminals who are blackmailing prominent social media personalities with scandalous “deepfake” videos. These videos use a deep learning model to splice a victim’s face onto an actor, creating highly realistic content that can be indistinguishable from the real thing. Your boss wants you to develop a method to efficiently detect these deepfakes from a huge data set of online videos. The method needs to be fast, and also run without needing GPU resources which are in short supply. Your challenges include gathering your datasets, training a support-vector machine (SVM) classifier to detect image artifacts left behind in deepfakes, and reporting your results to your superiors.

This project is designed for learning purposes and is not a complete, production-ready application or solution.

book resources

When you start your liveProject, you get full access to the following books for 90 days.

project author

Pavel Korshunov

Pavel Korshunov is a researcher at Idiap Research Institute, Switzerland, working on detection of audio-visual inconsistencies and Deepfakes. Previously, he worked on problems related to high dynamic range imaging, crowdsourcing, and visual privacy. He received PhD from National University of Singapore and MSc from St. Petersburg State University, Russia. He has over 70 research papers with several best paper awards and is a co-editor of JPEG XT standard.

prerequisites

This liveProject is for developers who know Python, the basics of machine learning, and the basics of processing image data. To begin this liveProject, you will need to be familiar with:

TOOLS

Basics of Jupyter Notebook
Basics of OpenCV
Basics of scikit-learn
Basics of Matplotlib
Basics of scikit-image

TECHNIQUES

Basic signal and image processing
Basic image feature and metrics computation
Basic understanding of how classification systems are evaluated
Basic understanding of classification with linear support vector machine

you will learn

In this liveProject, you’ll learn techniques and tools for image processing, face detection, and training a SVM classifier. These skills are easily transferable to common computer vision challenges you’ll face in industry.

Video and image processing
Binary classification and the evaluation of binary classifiers
Implementing SVM classification with scikit-learn
Reading videos and detecting faces with OpenCV
Facial image processing with scikit-image
Interpretation of evaluation results
Producing actionable reports

features

Self-paced: You choose the schedule and decide how much time to invest as you build your project.
Project roadmap: Each project is divided into several achievable steps.
Get Help: While within the liveProject platform, get help from other participants and our expert mentors.
Compare with others: For each step, compare your deliverable to the solutions by the author and other participants.
book resources: Get full access to select books for 90 days. Permanent access to excerpts from Manning products are also included, as well as references to other resources.