Chandigarh (National Times): In a significant leap forward for forensic science, researchers at Panjab University have developed an artificial intelligence-based facial reconstruction system capable of recreating human faces with up to 95% accuracy, using nothing more than the dimensions of the jaw and teeth.
This pioneering technology, now officially copyrighted by the Copyright Office of the Government of India, could revolutionize disaster victim identification (DVI) and forensic investigations, especially in scenarios where facial features are damaged or destroyed, such as plane crashes, bombings, or natural disasters.
The AI models were created by a multidisciplinary team led by Professor Kewal Krishan from the Department of Anthropology, with contributions from Professor Vishal Sharma of the Institute of Forensic Science and Criminology and Dr Arun K Garg of Dr. Harvansh Singh Judge Institute of Dental Sciences and Hospital. The research was further supported by six PhD scholars: Damini Siwan, Nandini Chitara, Ankita Guleria, Rakesh Meena, Akansha Rana, and Ayushi Srivastava.
Utilizing machine learning techniques, the models analyze preserved dental and mandibular structures to generate detailed facial predictions—even when a body is decomposed, mutilated, or fragmented. This offers a scientific pathway to identification in cases that were previously dependent on DNA or circumstantial evidence.
The work forms part of the doctoral research of Damini Siwan, who specializes in AI applications in forensic science, and Dr Nandini Chitara, who has since joined the Regional Forensic Science Laboratory in Panchkula, Haryana, as a Senior Scientific Assistant (Biology).
The AI tool is being hailed as a potential game-changer for forensic anthropology and could soon be implemented in national-level DVI protocols and complex criminal investigations.
