Researchers at NVIDIA, Aalto University, and MIT, have developed a deep-learning AI that can fix photos by just looking at examples of corrupted photos only, or in this case grainy images. Simply put, a neural network is shown example pairs of noisy and clean images, and then the AI learns how to make up the difference. Utilizing NVIDIA Tesla P100 GPUs with the cuDNN-accelerated TensorFlow deep learning framework, the team trained their system on 50,000 images in the ImageNet validation set. Continue reading for another video and more information.
“It is possible to learn to restore signals without ever observing clean ones, at performance sometimes exceeding training using clean exemplars. [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,” said the researchers in their paper.
