MUMBAI, India, Oct. 5 -- Intellectual Property India has published a patent application (202621099093 A) filed by Dr. D. Y. Patil Institute Of Technology, Pimpri, Pune -; and Dnyaan Prasad Global University, School Of Technology And Research, Pimpri, Pune - on August 17, 2026, for Real-Time Deepfake Detection System With Integrated Liveness Challenge Using Multimodal Artificial Intelligence.
Inventors include Ms. Jyotsna Barpute; Mayuri Nikade; Rajas Fegade; Asim Pathan; and Pranjal Mohite.
The application for the patent was published on October 02, 2026, under issue no. 40/2026.
Abstract: The invention relates to the auto-detection of deepfake media online/on-the-fly using AI methods. The invention takes the image/video/live feed for detection; it then applies various analysis methods to determine if the input is real/fake, it later applies the spatial analysis technique to detect any visual anomalies, temporal analysis to detect any irregular motions and Audio-Visual synchrony to measure the accuracy of lip synchronization. A liveness verification stage is designed for this system; simple challenge response activities is performed to verify whether a live user is faced with and protect against simulation or replay attack. The results of each module are fused into a final decision with confidence score and optional manipulated region map. This structure is designed to be used in a real-world environment and could be implemented on cloud based or edge-based platforms. It can be used for applications such as identity authentication, secure messaging, online exam, digital forensic. This invention will provide a very simple and effective way to ensure media authentication and confidence.
Disclaimer: Curated by HT Syndication.
