Deepfakes: ISI Researchers Develop Forensic Techniques to Identify Tampered Videos

June 21, 2019

By Amy Blumenthal

Computer scientists from the USC Information Sciences Institute (USC ISI), including Ekraam Sabir, Jiaxin Cheng, Ayush Jaiswal, Wael Abd-Almageed, Iacopo Masi, and Prem Natarajan, have developed a method that performs with 96 percent accuracy in identifying deepfakes when evaluated on large scale deepfake dataset.

The method works on various types of content manipulations known as deepfakes, faceswaps and face2face. At the point of publishing their paper, “Recurrent Convolutional Strategies for Face Manipulation Detection in Videos,” the authors said their detection method was ahead of the content manipulators who quickly modify as new detection methods arise. 

The paper was shared June 17 at the IEEE conference on Computer Vision and Pattern Recognition conference in Long Beach, California.

While previous methods of detecting deepfakes would often use “frame-by-frame” analysis of various aspects of a video, these prior methods, the USC ISI authors contend, are quite computationally heavy, take more time, and also have greater room for error. However, the newer tool developed by USC ISI, which was tested on over 1,000 videos, is less computationally intensive. It thus has the potential to scale and be used to automatically detect fakes that are uploaded in the millions of profiles on Facebook or other social media platforms in near real-time.

This effort, led by principal investigator Wael Abd-Almageed, a computer vision, facial recognition and biometrics expert, looks at a piece of video content as a whole.  The researchers used artificial intelligence to look for inconsistencies in the images through time, not just on a “frame-by- frame” basis. This is a key distinction says Abd-Almageed, as sometimes you cannot detect the manipulation on a frame-by- frame level, but by looking for facial motion inconsistencies.

To develop this first forensic tool, the USC ISI researchers used a two-step process. First, they input hundreds examples of verified videos of a person. Then they laid each video on top of one another. Then, using a deep learning algorithm known as a convolutional neural network, the researchers identified features and patterns in a person’s face, with specific attention to how the eyes close or how the mouth moves.  Once they had a model for an individual’s face and the movements surrounding their facial movements, they could develop a tool that compares a newly input video with the parameters of the previous models to determine if a piece of content was outside the norm and thus not authentic. One can imagine this working in the same way a biometric reader recognizes a face, retina scan or fingerprint.

“If you think deep fakes as they are now is a problem—think again. Deep fakes as they are now are just the tip of the iceberg and manipulated video using artificial intelligence methods will become a major source of misinformation,” Abd-Almageed says. One can imagine a world where everyone guards their video assets as much as they guard their bank PIN number.

The project is funded by the Defense Advanced Research Projects Agency (DARPA) MediFor program.