If you’re a fan of true crime, you probably know how often the wrong person gets picked in a police line-up. But guess what? Some smart psychologists are saying that the reason for this is the way police line-ups are done. They say that the current procedures don’t really help witnesses pick out the suspects effectively. But hey, there’s hope! New AI technology might just be the answer to this problem.
So, here’s the deal. In the movies, you see victims choosing suspects from a group of people paraded around a room, right? Well, that’s not how it usually goes down in real life. Instead, victims only get to see still photos of the suspects, which doesn’t give them a good look at who might have done it.
But hold on, there’s a new system in the works. Researchers at the University of Birmingham have come up with a cool interactive system that uses 3D models to show the suspects from different angles – just like the victim saw them. And guess what? It works! The researchers found that this interactive viewing thingy makes it 42% more likely for the witness to pick the right suspect. And get this, it’s even better than video line-ups where the suspects are filmed turning around – the technology improved selection accuracy by 20%.
Now, here’s the real talk. Prof Heather Flowe, who’s been working on this project for over nine years, says that the current line-up methods used by police forces in the UK and USA are just not cutting it. In the US, the suspects’ photos don’t even have to match – they can range from driver’s license photos to people in prison uniforms. It’s a total mess, according to Flowe. She says it’s not fair at all.
But wait, there’s more! Flowe and her team are working on some cool tricks to help jog the witness’s memory. They’re testing ways to include dynamic facial movements, emotional expression, changes in lighting, and even accessories like masks using AI. The idea is that the more details the witness gets, the better their choice will be.
And here’s the kicker. Flowe is all about getting this new line-up technology out there. Her team is talking to the US and UK police forces to start testing it in the field. She says it’s about time the police start using better tech, especially since line-ups haven’t really changed in 100 years.
So, how does this new tech work? It turns video into an interactive 3D image that witnesses can click and drag to see the suspects from different angles. It’s like the ultimate line-up procedure, according to Flowe. She’s all about giving the witness the chance to see the suspect just like they did in real life.
Flowe brings up the case of Andrew Malkinson, who was wrongfully convicted in 2003 because the victim mistakenly chose him in a line-up, even though he didn’t really match the description of the actual perpetrator. She says there are a lot of cases like this where we just haven’t thought about how the witness saw the suspect.
And get this, Flowe even wrote a paper about the Ted Bundy case. One of the survivors saw Bundy from the side during the attack, but later had to pick his photo from a front-facing line-up. Bundy’s defence said her choice must have been influenced by seeing his image in newspapers. Flowe is pretty frustrated about this and says we should show the witness what they actually saw.
Looking ahead, Flowe hopes that technology will bring static photos of suspects to life. She’s working on a project to see how showing witnesses different emotional expressions on suspects’ faces during line-ups could make a difference. And guess what? They’ve got a research grant to work on this with experts from different parts of the world.
But here’s the big question. How do you make suspects act out emotions for a line-up? Well, that’s where the AI comes in. Flowe’s team is testing the use of AI to create line-ups with realistic facial expressions. But she knows they need to be careful because AI could mess up and show false emotions.
And here’s the really cool part. In the future, AI could even be used to superimpose suspects into the scene with all the evidence. Flowe says there’s a strong theory to support this idea.
+ There are no comments
Add yours