MUMBAI, India, Jan. 23 -- Intellectual Property India has published a patent application (202511127234 A) filed by Swami Rama Himalayan University, Dehradun, Uttarakhand, on Dec. 16, 2025, for 'a system for attributing deepfake videos via temporal signature analysis.'
Inventor(s) include Dr. Suman Pant; Ms. Shivani Pant; Dr. Anupama Mishra; Dr. Ashutosh Bhatt; and Dr. Pramod Kumar.
The application for the patent was published on Jan. 23, under issue no. 04/2026.
According to the abstract released by the Intellectual Property India: "This invention relates to the field of digital video forensics and deepfake detection. It discloses a system comprising an input video acquisition module that extracts metadata including frame rate and encoding details, a frame extraction module that decomposes video content into timestamped frames, and a temporal feature extraction module that computes features such as the FTDI, METD, ITV, LFTD, and MVTC. The temporal signature generator module aggregates these features into a unified vector using advanced neural networks like CNN-LSTM, transformer, or temporal graph neural networks. A source identification engine compares this vector with reference vectors in a dynamic signature database and an output module presents an attribution result with a quantified confidence measure."
Disclaimer: Curated by HT Syndication.
