Researcher at the Indian Institute of Technology, Ropar in Punjab and Monash University, Australia, have developed a unique detector called “FakeBuster” to identify impersonators attending a virtual conference without anyone’s knowledge. It can also find out faces that have been manipulated on social media to defame someone or to make a joke.
In the current pandemic scenario where most of the official meetings and work is done online, this standalone solution is intended to enable a user (organizer) to see if someone else’s video is being tampered with or faked during one Video conferencing. This means that technology will find out if a scammer is attending a webinar or virtual meeting on behalf of one of your coworkers by transforming their image with their own.
“Sophisticated artificial intelligence techniques have led to a dramatic increase in the manipulation of media content. Such techniques evolve and become more realistic. This makes it difficult to identify, which could have far-reaching effects on safety, ”said Dr. Abhinav Dhall, one of the most important members of a four-person team that developed the “FakeBuster”. “The tool achieved an accuracy of over 90 percent,” assures Dr. Dhall. The other three members are associate Prof. Ramanathan Subramanian and two students Vineet Mehta and Parul Gupta.
An article on this technique ‘FakeBuster: A. DeepFakes The video conferencing scenario detection tool was unveiled last month at the 26th International Conference on Intelligent User Interfaces in the United States.
Dr. Dhall said that the use of tampered media content in the distribution of fake news, pornography and other such content online has been widely seen with great impact. He said such manipulations recently found their way into video calling platforms through spoofing tools based on the transmission of facial expressions. These fake facial expressions are often convincing to the human eye and can have serious consequences. These real-time mimicked graphics (videos), known as “deepfakes”, can even be used in online exams and job interviews.
This software platform is independent of video conferencing solutions and has been tested with Zoom and Skype Applications.
The deepfake detection tool “FakeBuster” works in both online and offline mode. As the device currently only connects to laptops and desktops, “we want to make the network smaller and lighter so that it can run on cell phones / devices,” said Associate Professor Subramanian. He said the team is working on using the device to detect fake audios as well.
The team claims that this software platform ‘FakeBuster’ is one of the first tools to detect fraudsters during live video conferencing using DeepFake detection technology. The device has already been tested and will soon be launched.