As concerns regarding financial safety and fraud involving manipulated videos escalate, Professor Saerom Park’s team in the Department of Industrial Engineering at UNIST has developed an innovative deepfake prevention technology that emphasizes proactive measures rather than reactive ones. The team was awarded the Excellence Award (FSI President’s Award) at the ‘8th Financial Securities Institute (FSI) Thesis Contest‘ held in September 2024.
Professor Park’s team has revolutionized deepfake prevention by introducing a method that advances beyond traditional noise addition techniques. Their technology effectively blocks deepfake attacks while preserving the quality of the original image to a greater degree.
While existing technologies apply noise uniformly across an entire image, Professor Park’s approach selectively adds noise only to specific areas targeted by deepfake generation models. This innovation minimizes the influence of deepfake attacks and significantly reduces the visual degradation of the original image.
The implications of this research extend beyond technical advancements, as it offers a practical solution to a pressing issue within the financial sector. The team’s findings represent a significant technological leap that can help avert potential threats to financial integrity.
“This study will greatly contribute to creating a safe environment for financial transactions,” said JoonHyuk Baek, a participating student. “We will continue to strive towards developing effective solutions to protect financial consumers.”
Meanwhile, the Financial Services Commission (FSC) Chairman’s Award, the grand prize of this contest, was awarded to the team from the Hana Institute of Technology (HIT). The NAVER Cloud Team received the Financial Supervisory Service (FSS) President’s Award, which is also known as the Best Excellence Award.
