Nvidia and scientists teach AI to remove graininess from photos

Researchers from MIT, Aalto University and Nvidia cleans up the noise from pictures

Scientists have developed an artificial intelligence system that can automatically remove noise, specks, and other distortions from pictures.

The technology, called  Noise2Noise AI, was developed by researchers from Nvidia, Aalto University in Finland, and MIT. The researchers used 50,000 pictures, as well as MRI scans and computer-generated noisy images, to train the system. According to the research paper, the AI can remove enough noise to make images usable again without ever seeing a clean image.

Nvidia Tesla P100 GPUs with the cuDNN-accelerated TensorFlow deep learning framework were used to train up the system.

The technology has been trained to remove noise without needing to understand what a clean image looks like, which until now, such AI work has focused on training a neural network to restore images by showing example pairs of noisy and clean images.

"It is possible to learn to restore signals without ever observing clean ones, at performance sometimes exceeding training using clean exemplars," the researchers said.

Advertisement
Advertisement - Article continues below
Advertisement - Article continues below

They added that the neural network is "on par with state-of-the-art methods that make use of clean examples using precisely the same training methodology, and often without appreciable drawbacks in training time or performance".

The system was tested using three different datasets to validate the neural network.

Scientists said there were several real-world situations where obtaining clean training data is difficult: low-light photography (e.g., astronomical imaging), physically-based image synthesis, and magnetic resonance imaging.

"Our proof-of-concept demonstrations point the way to significant potential benefits in these applications by removing the need for potentially strenuous collection of clean data,"  said the paper's authors. "Of course, there is no free lunch we cannot learn to pick up features that are not there in the input data but this applies equally to training with clean targets." 

Featured Resources

Transform the operator experience with enhanced automation & analytics

Bring networking into the digital era

Download now

Artificially intelligent data centres

How the C-Suite is embracing continuous change to drive value

Download now

Deliver secure automated multicloud for containers with Red Hat and Juniper

Learn how to get started with the multicloud enabler from Red Hat and Juniper

Download now

Get the best out of your workforce

7 steps to unleashing their true potential with robotic process automation

Download now
Advertisement

Most Popular

Visit/operating-systems/microsoft-windows/354297/this-exploit-could-give-users-free-windows-7-updates
Microsoft Windows

This exploit could give users free Windows 7 updates beyond 2020

9 Dec 2019
Visit/security/vulnerability/354309/patch-issued-for-critical-windows-bug
vulnerability

Patch issued for critical Windows bug

11 Dec 2019
Visit/cloud/microsoft-azure/354230/microsoft-not-amazon-is-going-to-win-the-cloud-wars
Microsoft Azure

Microsoft, not Amazon, is going to win the cloud wars

30 Nov 2019
Visit/data-insights/big-data/354311/google-reveals-uks-most-searched-for-terms-in-2019
big data

Google reveals UK’s most searched for terms in 2019

11 Dec 2019