Science

AI app transforms black and white photos into colour


Now you can turn historical images and old family portraits into COLOUR photos with a free AI app that uses machine learning to transform black and white images in seconds

  • ColouriseSG can turn old black and white photographs into images with full colour in seconds 
  • The online tool was developed to provide Singapore with a viable and easy colourisation tool  
  • It has no way of knowing what the original colours should be but is designed to give a best estimate  

Advertisement

An online tool has been unveiled which is capable of bringing black and white photographs to life instantaneously by adding colour to them using artificial intelligence. 

Colourisation of old images is a normally time consuming process which requires specialist training and expensive software.

The tool, ColouriseSG, is able to do it for free from only a single digital image and works on iconic historical photographs and old family portraits. 

Try it for yourself here or via the interactive tool below. 

London, England: Big Ben and the Houses of Parliament from across the Thames in London. The black and white image was instantly converted to colour using the free tool 

Migrant Mother image taken by Dorothea Lange. The woman – Florence Owens Thompson – lived between 1895 and 1964 and was snapped on silver print in 1936

It is trained on a back catalogue of old images and uses machine learning to guess what it thinks the image would have looked like in colour. 

‘The purpose of colourisation is to generate an image with colours that are plausible,’ the tool’s developers claim.  

‘It by no means guarantees that the colourised image is an accurate representation of the actual snapshot in time.’

It is also more adept at colourising images of human subjects with natural landscape and can struggle with more complex pictures. 

The free-to-use tool was developed to provide an accurate way for people from Singapore to edit their monochrome images. 

Existing software, such as Algorithmia, are trained using 1.3 million images from ImageNet, a database of photographs developed in the US by researchers at Stanford University and Princeton University.

The idea of ColouriseSG was to provide a large enough data-set to be relevant to the residents of Singapore. 

Tech firm NVIDIA have released several different image editing tools that are powered by artificial intelligence. 

US Marines of the 28th Regiment, 5th Division, raise the American flag atop Mount Suribachi, Iwo Jima, in this image taken February 23, 1945. It was strategically located only 660 miles from Tokyo, the Pacific island became the site of one of the bloodiest, most famous battles of World War II against Japan

The frilly, fluffy look which Bohan showed for the evening line of Dior’s Spring Collection is typified in this model in 1900 

WHAT IS MACHINE LEARNING?

Artificial intelligence systems rely on neural networks, which try to simulate the way the brain works in order to learn.

These networks can be trained to recognise patterns in information – including speech, text data, or visual images – and are the basis for a large number of the developments in AI.

They use input from the digital world to learn, with practical applications like Google’s language translation services, Facebook’s facial recognition software and Snapchat’s image altering live filters.

But the process of inputting this data can be extremely time consuming, and is limited to one type of knowledge.

Its latest release is software which uses deep-learning to elevate even the roughest sketches into works of art.

The new program, dubbed GauGAN, after famous French impressionist Paul Gaugin, uses a tool called generative adversarial networks (GAN) to interpret simple lines and convert them into hyper-realistic images.

Its application could help professionals across a range of disciplines such as architecture and urban planning render images and visualisations faster and with greater accuracy, according to the company.

‘It’s much easier to brainstorm designs with simple sketches, and this technology is able to convert sketches into highly realistic images,’ said Bryan Catanzaro, vice president of applied deep learning research at NVIDIA.

What separates GauGAN from other rendering software, beyond its seemingly simple interface and application, is just how the tool was developed.

According to NVIDIA, GauGAN was trained to mimic different types of landscapes by employing the use of deep-learning software to pore over 1 million different images. 

Tech firm NVIDIA have released several different image editing tools that are powered by artificial intelligence.  Its latest release is software which uses deep-learning to elevate even the roughest sketches into works of art. The new program, dubbed GauGAN, after famous French impressionist Paul Gaugin, uses a tool called generative adversarial networks (GAN) to interpret simple lines and convert them into hyper-realistic images

Tech firm NVIDIA have released several different image editing tools that are powered by artificial intelligence.  Its latest release is software which uses deep-learning to elevate even the roughest sketches into works of art. The new program, dubbed GauGAN, after famous French impressionist Paul Gaugin, uses a tool called generative adversarial networks (GAN) to interpret simple lines and convert them into hyper-realistic images





READ SOURCE

Leave a Reply

This website uses cookies. By continuing to use this site, you accept our use of cookies.