In January 2024, the Delhi Police cracked a “blind case”, with the help of AI. A 30-member team used AI technology to reconstruct the facial features of an unidentified man whose face was damaged beyond recognition. The AI-generated image, showing the man with his eyes open, was a breakthrough.
The police even altered the background and uploaded it to the Crime and Criminal Tracking Network (CCTN) website, hoping for a lead. That lead came when the deceased’s brother saw the image, and reached out to the police to identify the victim as Hitendra.
This is just the beginning.
Recently, Dainik Bhaskar and the Rajasthan Police collaborated to use AI to solve cases of children gone missing. Joining this effort is Sahid NK, a young graphic designer who brings a creative touch to the project.
In an exclusive interview with AIM, Sahid shared, “I get old, worn-out photos where the faces are so faded that it’s hard to even recognise them. We have to be imaginative and use the little details we have to recreate a face from the past.”
Sahid mentioned that he relies on AI-powered tools to bring these faces back to life. “I use models trained to understand intricate facial features. It’s a blend of creativity, skill, and technology,” he said.
Among his go-to tools are Freepic’s Picasso and the sophisticated Illusion Diffusion, a generative AI tool.
Sahid draws inspiration from Kenya, where a similar AI initiative helped find 496 missing children in just four years. In India, the challenge is even greater as the number of missing children had risen to 47,000 by the end of 2023.
But with AI’s growing role, there’s still hope for the families of these kids.
More of AI Helping Solve Crime in the Country
Meanwhile, a Gurugram-based startup, Staqu Technologies, is blending advanced AI solutions to solve crime cases in the country. It has partnered with several state governments, including Uttar Pradesh and Punjab, to enhance public safety and security.
While its primary focus is supporting police and security forces, only 30-35% of its revenue comes from government contracts. A majority of the revenue is generated from the private sector, where it uses video analytics for various security and operational use cases.
This includes intrusion detection, fire detection, SOP compliance in manufacturing, retail industry analytics like tracking queue lengths, customer churn, and more.
A standout innovation from Staqu is Crime GPT, an AI tool created by the company, which helped the UP Police catch criminals.
Crime GPT can quickly extract information about individuals through both written and spoken inquiries. Its features, including facial and vocal recognition, along with the analysis of criminal networks, promise to streamline investigative processes.
By tapping into digital criminal databases, Crime GPT equips police departments with insights, facilitating specific details about their queries.
Crime GPT is an extended version of Staqu Technologies’ tool Trinetra. It is renowned for its proficiency in tracking criminals via facial and vocal cues. With Trinetra, the UP Police have a database with information on over 900,000 criminals.
Every day, the company identifies over 400 criminals. Over the past six or seven years, it has identified more than 30,000 criminals. This represents a significant value proposition for the Indian ecosystem, given the high population density, which makes crime and evasion relatively easier.
What’s Happening in Other Countries?
India isn’t the only country adopting AI for crime-solving. In the US, UK, Japan, and Singapore, law enforcement agencies are using AI to tackle various criminal activities.
In one of the reports, Armando Aguilar, the Miami assistant police chief noted, “Before AI, the police department was only able to arrest suspects in 45% of murders and less than 38% of violent crimes.”
After implementing AI in 2023, the clearance rate for murders jumped to 68%, and increased to 58% for violent crimes.
AI algorithms are revolutionising how law enforcement processes video and image data, with capabilities like facial recognition, weapon identification, and event detection.
In Malaysia, researchers are developing AI software specifically for CCTV cameras to reduce street crime, with the ability to detect weapons, aggressive behaviour, and alert authorities in real-time.
