MUMBAI, India, Aug. 10 -- Intellectual Property India has published a patent application (202641091334 A) filed by Bgs College Of Engineering And Technology Bgscet on July 28, 2026, for Artificial Intelligence-Based Forensic Suspect Image Generation Method And System.
Inventors include Ravikumar Guralamata Krishnegowda; Govindaiah Thimma Raju; Parvathi Chikkanna; Mamatha Jadav Venkatarao; Ashutosh Kumar Jha Sanjay; Bhadreesh Singanapalya Mahesh; Charan Sri Hari Prakash; and Naveen Sreekanth.
The application for the patent was published on July 31, 2026, under issue no. 31/2026.
Abstract: The present disclosure addresses limitations associated with conventional forensic sketching and existing artificial intelligence-based suspect image generation techniques, including inaccurate facial reconstruction, limited attribute controllability, and inconsistent refinement. The present disclosure provides an artificial intelligence-enabled computer-implemented method and a suspect image generation system (202) configured to generate photorealistic suspect facial images from structured eyewitness descriptions. The system comprises an input and data acquisition module (210), a preprocessing module (212) configured to generate standardized semantic representations, an artificial intelligence image generation engine (214) employing a diffusion-based text-to-image model, a post-processing module (216), an output and visualization module (218), a model repository (224), and a feedback and iterative refinement module (226) configured to selectively regenerate modified facial attributes while preserving previously accepted facial characteristics.
Disclaimer: Curated by HT Syndication.
