This could be an important step in exposing an impostor who participated in a virtual meeting without anyone knowing or representing someone. The Indian Institute of Technology (IIT) Roper in Punjab, India and Monash University in Australia Researchers have developed a unique detector FakeBuster. It not only exposes imposters, but also finds faces that have been manipulated on social media to slander others or make jokes. FakeBuster can be seen as the way forward, especially in the current era when most work and formal meetings are conducted online.
The latest technology can help organizers of online meetings or seminars to detect whether participants’ videos have been tampered with during the interaction. FakeBuster will detect whether someone is attending a meeting on behalf of a colleague by deforming its own image.
At the 26th International Conference on Intelligent User Interfaces held in the United States in April, a paper titled “FakeBuster: DeepFakes Detection Tool for Video Conference Scenes” was published. Researchers say the software has been tested by Zoom and Skype. Its creator said that more importantly, the tool can work in both online and offline modes.
Dr. Abhinav Dhall, one of the key members of the four-person team that developed FakeBuster, said in a statement that advanced AI technology has greatly promoted the manipulation of media content and has continued to evolve and become more realistic. He said: “The tool has reached over 90% accuracy.”
The other three members of the team include Associate Professor Ramanathan Subramanian and two students, Vineet Mehta and Parul Gupta.
Subramanian said that the device can be connected to laptops and desktops, adding that their goal is to “make the network smaller and lighter, so that it can also run on mobile phones/devices.” Not only that, the professor added Said that the team is still working hard to use the device to detect false audio.
Dr. Dhall added that the use of manipulated media content to spread fake news, pornography and other such online content has caused widespread repercussions. He said that these manipulations have also entered the video call platform through deception tools based on facial expressions. Researchers say these fake facial expressions are usually convincing and can have serious effects. What people worry about now is that these real-time imitated visual effects (called deepfakes) can even be used in online exams and job interviews.
Dr. Dhall’s team claims that FakeBuster uses DeepFake detection technology and is one of the earliest tools to detect imposters during real-time video conferences. It is expected to be available soon.