Researchers at NVIDIA have announced a new AI program that can remove noise and grain from images with very accurate results.
What’s most interesting is that researchers trained the AI using two images with different noise patterns, rather than giving it a noisy and a clean image.
‘It is possible to learn to restore signals without ever observing clean ones, at performance sometimes exceeding training using clean exemplars,’ the research team said in their blog post announcing the breakthrough. ‘[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.’
Using NVIDIA Tesla P100 GPUs with the cuDNN-accelerated TensorFlow deep learning framework, the team trained its system on 50,000 images in the ImageNet validation set.