Which World Cup player do YOU look like? Find out with this AI tool that reveals your football doppelganger
- New AI-powered tool matches your face with a professional footballer player
- Advanced facial recognition takes 128 measurements from each face
- We tested the software to see how accurate it was – and had surprising results
Are you more of a Ronaldo or a Kane?
A new AI-powered tool, which uses facial recognition technology to find your football doppelganger, will help you find out.
The artificial Intelligence will attempt to match your likeness with one of the 736 professional sportsmen currently competing in the World Cup in Russia.
Fancy a go? You can test the hilarious tool below – or via this link.
To test the ‘Find Your World Cup Twin’ tool, hit the Upload Image button at the bottom of the screen.
On desktop computers, this will pop-up a dialogue box where users can upload existing images in JPEG or PNG formats.
When viewed on a mobile device, tapping the Upload Image button will allow users to take photos using the built-in camera, as well as upload existing pictures.
MailOnline put the system through its paces with a mixture of celebrities, team members, and even a few World Cup footballers – to see if the system would recognise them correctly.
Overall, the results were pretty good.
The tool wasn’t tripped up by the professional footballers and correctly matched photos of Lionel Messi and Cristiano Ronaldo with the correct World Cup players.
Meanwhile, Ed Sheeran was matched to German footballer Toni Kroos.
A new AI-powered has been released that uses facial recognition to reveal your football doppelganger. It matched British comedian Sacha Baron Cohen with Uruguay national team defender Diego Godin with a 61 per cent likeness
While some matches were accurate, the system also produced some amusing results. Prince William apparently looks most like Egypt and Aston Villa player Ahmed Elmohamady
The tool says that Daniel Craig and German winger Julian Draxler are close matches, with a 52 per cent likeness
Star Wars and Pacific Rim: Uprising actor John Boyega’s lookalike was Nigerian goalkeeper Daniel Akpeyi
According to the facial recognition software, there is a 56 per cent resemblance between the Galway Girl singer and the Real Madrid midfielder.
Unfortunately, the AI is not 100 per cent accurate and has made some amusing matches.
Kim Kardashian West was matched with 33-year-old Senegalese footballer, Khadim N’Diaye.
Find Your World Cup Twin believes there was a 37 per cent resemblance between Kardashian and N’Diaye.
Hollywood star Shia LaBeouf matched closest with Portugal left back Mário Rui, according to the AI
The AI tool, dubbed Find Your World Cup Twin, was developed by a team working at Norwegian news organisation, VG. MailOnline science and technology journalist Aaron Brown matched with Croatia and Barcelona midfielder Ivan Rakitic with a 53 per cent accuracy
Actor Daniel Radcliffe shares a 57 per cent likeness with Iranian player Alireza Jahanbakhsh, according to the software
The system has already been trained to pinpoint 128 measurements on each face it detects within a photograph. MailOnline health reporter Sam Blanchard matched with Poland player Lukasz Piszczek
The AI tool, dubbed Find Your World Cup Twin, was developed by a team of developers working at Norwegian news organisation, VG.
The publisher used an open-source facial recognition library developed by software engineer Adam Geitgey as the foundation for its new tool.
Coded in the Python programming language, the system has already been trained to pinpoint 128 measurements on each face it detects within a photograph.
These facial measurements can be used to match the likeness between two people.
The AI will attempt to match your likeness with one of the 736 professional sportsmen competing in the 2018 World Cup. It is not always accurate, suggesting Kim Kardashian’s closest lookalike is Senegal goalkeeper Khadim N’Diaye
Chris Lawrence, MailOnline head of social media and communities, shared a likeness with Tunisian player Fakhreddine Ben Youssef
Australian midfielder Mark Milligan bears a striking resemblance to Meghan Markle, according to the Find Your World Cup Twin AI
Although the library created by Geitgey was trained on a large dataset, the developers had to tailor the tool to specifically recognise World Cup team photos.
For the final tool, the VG team also incorporated technology from OpenCV (Open Source Computer Vision Library) – an open-source database which contains 2,500 free-to-use algorithms optimised for machine learning and computer vision.
Machine learning is a process that enables a computer to learn from data and draw its own conclusions, while computer vision describes any practice that enables machines to gain a high-level understanding from digital images or videos.
The AI is not 100 per cent accurate and has made some amusing matches. MailOnline science and technology journalist Harry Pettit was paired with Poland player Jacek Goralski
Ed Sheeran was matched to German footballer Toni Kroos. According to the facial recognition software, there is a 56 per cent resemblance between the Galway Girl singer and the Real Madrid midfielder
The Find Your World Cup Twin app wasn’t tripped up by the professional footballers and correctly matched photos of Lionel Messi and Cristiano Ronaldo with the correct World Cup players
‘Tying it all together are several image- and request processing programs of our own, which all run in Google’s App Engine to scale geographically,’ Einar Otto Stangvik, one of the developers behind the face-matching service, told MailOnline.
The Find Your World Cup Twin team decided to soft launch their tool on Facebook to see how it handled images uploaded by the public.
Otto Stangvik said: ‘Through the night and wee hours of the morning, close to 30,000 images had been run through it.’
This allowed the team to make any final improvements to the system before rolling it out to a larger audience.
HOW DOES FACIAL RECOGNITION TECHNOLOGY WORK?
Facial recognition software works by matching real time images to a previous photograph of a person.
Each face has approximately 80 unique nodal points across the eyes, nose, cheeky and mouth which distinguish one person from another.
A digital video camera measures the distance between various points on the human face, such as the width of the nose, depth of the eye sockets, distance between the eyes and shape of the jawline.
A smart surveillance system (pictured) that can scan 2 billion faces within seconds has been revealed in China. The system connects to millions of CCTV cameras and uses artificial intelligence to pick out targets
This produces a unique numerical code that can then be linked with a matching code gleaned from a previous photograph.
A facial recognition system used by officials in China connects to millions of CCTV cameras and uses artificial intelligence to pick out targets.
Experts believe that facial recognition technology will soon overtake fingerprint technology as the most effective way to identify people.
Source: Read Full Article