The Ludhiana Police has used video quality enhancement algorithms to apply facial recognition and track down criminals in the city. While this is not the first instance of a police department using facial recognition to track offenders, the latest case was based on footage from a low resolution camera, in a surveillance network that operated on low bandwidth networks.
The algorithm, which was provided by Indian artificial intelligence-based video analytics firm Staqu, offered enhancement of low-quality camera footage to enhance the quality of video imagery captured via closed circuit television (CCTV) cameras. Using this footage, the Ludhiana Police could track down criminals who have since been captured.
In August 2020, the Delhi Police used similar video analytics and facial recognition technologies to track down three criminals – using a database of criminals and a network of over 150 CCTV cameras to track them down.
Delhi Police’s facial recognition platform initially used facial recognition technology offered by Indian startup Innefu Labs. Startups such as Innefu, as well as Staqu, offer analytics, quality augmentation and related facial recognition technologies to various law enforcement bodies in order to cut down on criminal activities.
However, legal experts have frequently stated that with no existing legal framework governing the use of facial recognition and personal data in India, such technologies leave many questions to be answered.
India’s upcoming Data Protection Bill, which is yet to be passed, has sought to lay down guidelines regarding the use of facial recognition by government and private organizations in India. Under these guidelines, use of personal data for facial recognition or video surveillance could be disallowed without prior consent – barring exceptional circumstances such as national security, or compliance with a court order.
Under this ambit, technologies such as Staqu’s Jarvis video analytics platform, which helped the Ludhiana Police track criminals down, have been improving their offerings with the use of artificial intelligence – combined with face databases and camera feeds.
Atul Rai, co-founder and chief executive of Staqu, said in a statement that the company’s new algorithms, which can also facilitate automatic number plate recognition for tracking down law-offending vehicles, are “robust, solutions-driven, and in sync with infrastructural challenges in different domains.”