AI-Based 3D Suspect Modeling for Enhanced Identification

This paper presents a novel AI-based 3D suspect modeling system aimed at overcoming the limitations of traditional manual sketch-based methods in law enforcement. By integrating Natural Language Processing (NLP), speech-to-text APIs, and 3D modeling technologies, the proposed system translates verbal descriptions into lifelike 3D facial models. This approach significantly enhances the accuracy and efficiency of suspect identification while streamlining investigative workflows. The system architecture features a user-friendly interface, a secure backend for data processing, and APIs for transcription, attribute extraction, and 3D rendering. The modeling system is tested on various cases, and the results demonstrated that it generates 3D facial models with approximately 87% accuracy in less than one second. These test cases highlight the system's efficiency and performance, establishing its effectiveness as an automated sketch-based method for law enforcement.