Get rid of those photobombers! Software removes distracting faces, litter and even passing cars from pictures

  • Scientists at Princeton University and Adobe have created new software that identifies unwanted objects and clutter in photos and removes them 
  • It can remove faces of photobombers that might ruin a prized moment
  • The one-click system can also remove litter, signs and passing cars 

It is a prank that can turn a photograph into a viral sensation, but it can also ruin someone's big moment.

However, it may soon be possible to remove those annoying photobombing strangers from images with new software being developed by computer scientists.

The one-click system can also remove other distracting objects that often spoil or clutter up images, such as litter, signs or passing cars.

Photobombs can ruin pictures - although sometimes it can make them better - but computer scientists have developed new software that can identify faces that appear not to be part of the main composition and remove them. A picture of actor Jared Leto photobombing Oscar winning producer James Skotchdopole is shown

Photobombs can ruin pictures - although sometimes it can make them better - but computer scientists have developed new software that can identify faces that appear not to be part of the main composition and remove them. A picture of actor Jared Leto photobombing Oscar winning producer James Skotchdopole is shown

The software could open up an element of photo-editing that has mainly been restricted to those with the technical know-how and skill to manually 'Photoshop' unwanted items from pictures.

The computer scientists, who have been working with software firm Adobe, said their prototype can detect cars in a digital image and automatically remove them if they appear to be entering the frame.

Ohad Fried, a graduate student in computer science at Princeton University who has been leading the project, said it can also detect unwanted faces that appear out of place in the rest of the image.

Computer scientists trained their software to identify features in photographs that are distracting and unwanted so it can remove them, like the labels in the top image and the sign in the bottom image

Computer scientists trained their software to identify features in photographs that are distracting and unwanted so it can remove them, like the labels in the top image and the sign in the bottom image

NEVER LET WINDOWS OR RAILINGS RUIN YOUR HOLIDAY SNAPS AGAIN

They are perhaps the most common cause of ruined holiday snaps, but computer scientist have now created a way to remove unsightly reflections and wire fences from images.

The software, created by researchers at Google and the Massachusetts Institute of Technology (MIT), is able to strip out reflections that often obscure pictures when taken through glass.

It can also remove dust and raindrops that can also ruin images taken through glass while also removing distraction, blurry fences from pictures.

The technology could prove particularly useful for those wanting to take pictures through windows of the view from the top of tall buildings or attractions like the Millennium Wheel in London. 

Images through fences at zoos or from the top of the Eiffel tower could also be stripped of the unsightly blurred wire that often ruins treasured snaps. 

However, the researchers behind the software could also prove useful for anyone also wanting to get a clear picture of something in a reflection. 

Police, for example, might be able to see who was in a room from an image taken looking out of a window. 

He said: 'If the face is large and in the centre of the photo, we probably don't want to remove it. But if it is coming in from the side, it might be a photobomb.'

They created the programme by identifying distracting elements in more than 1,073 images.

They also used 5,000 images from an experimental Adobe app for the iPhone called Fixel, which allows photographers to retouch and upload images.

This allowed them to compare the before and after pictures to see which elements the photographers had removed.

They used these to train the software to spot unwanted features in photographs.

Specific arrangements of colours and shapes were found to be most associated with certain distracting elements. This allowed the software to identify and eliminate these.

Even members of the Royal family cannot resist a photobomb, as the image of two members of the New Zealand Olympic Committee, with Prince Harry in the background reveals. With the software the cheeky Prince could have been automatically edited out of the image

Even members of the Royal family cannot resist a photobomb, as the image of two members of the New Zealand Olympic Committee, with Prince Harry in the background reveals. With the software the cheeky Prince could have been automatically edited out of the image

The software can identify faces that may appear out of place in the image and can remove them automatically from the picture. However, when large areas are obscured, like in the photobomb above, it may be difficult to save some pictures

The software can identify faces that may appear out of place in the image and can remove them automatically from the picture. However, when large areas are obscured, like in the photobomb above, it may be difficult to save some pictures

The researchers hope their software, which they presented at the 2015 Computer Vision and Pattern Recognition Conference in Boston, could help people using social networking sites upload better quality images.

Their software could automatically identify a distracting sign, litter or photobomb and remove it.

For people who have had their wedding or holiday photographs ruined by someone peeking into the image, it could be a big help.

In their paper, the team wrote: 'Last year Facebook reported that people were uploading photos at an average rate of 4,000 images per second.

'The overwhelming majority of these pictures are casual — they effectively chronicle a moment without much work on the part of the photographer.'

Mr Fried added: 'In the future, it might become accurate enough to be used in a photo editing product. Right now, it is a very interesting research question.'

 

No comments have so far been submitted. Why not be the first to send us your thoughts, or debate this issue live on our message boards.

By posting your comment you agree to our house rules.

Who is this week's top commenter? Find out now