MUMBAI, India, Jan. 2 -- Intellectual Property India has published a patent application (202541123342 A) filed by Vardhaman College Of Engineering, Hyderabad, Telangana, on Dec. 7, 2025, for 'generative adversarial network (gan)-based method for high-fidelity image super-resolution.'
Inventor(s) include Dr. B K Madhavi; Dr. Muni Sekhar Velpuru; Ms. Ch Dhanalaxmi; Ms. Sumaiya Sk; Dr. Srinivasulu Gogula; and Dr. Ganesh Bhaiyya Regulwar.
The application for the patent was published on Jan. 2, under issue no. 01/2026.
According to the abstract released by the Intellectual Property India: "The invention provides a GAN-based high-fidelity image super-resolution method that reconstructs fine-grained details from low-resolution inputs using a multi-stage generator, a perceptually guided discriminator, and an adaptive loss fusion mechanism. The system enhances texture accuracy and minimizes noise by combining pixel loss, adversarial loss, perceptual feature loss, and frequency-domain consistency loss. The invention introduces an attention-driven refinement module and a progressive residual learning pipeline to preserve structural integrity while enhancing edges, textures, and color distribution. The method supports static images, video frames, and multi-modal inputs, delivering superior sharpness, naturalness, and visual coherence compared to conventional interpolation and deep CNN-based techniques, making it suitable for medical imaging, satellite imagery, security footage, digital restoration, and consumer media enhancement applications."
Disclaimer: Curated by HT Syndication.
