While modern digital cameras have made significant strides in shooting cleaner images at high ISOs, many photographers still do battle with image noise on a regular basis.
Chip maker NVIDIA has just revealed a new technique, based on deep learning, that can quickly dispatch image noise.
As NVIDIA explains, typical deep learning approaches have required training a neural network to recognize when a clean end state image should look like based on a series of noisy images. Armed with this information, the network can then take a fresh, noisy image and remove the noise. But NVIDIA’s new technique works without needing to be feed samples of noisy images.
“It is possible to learn to restore signals without ever observing clean ones, at performance sometimes exceeding training using clean exemplars,” the researchers stated in their paper. “[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.”
Beyond improving low light images, NVIDIA’s technique can be used to improve MRI images, astronomical images and more. NVIDIA itself isn’t really in the software business–they are pioneering AI techniques in large part to drive the use of their GPU chips, which are the backbone of AI processing. But their work undoubtedly has significant implications for photography–and photographers–in the years to come.
Just consider what the company has accomplished to date. Prior to the paper on de-noising, they have demonstrated the ability to use AI to turn normal video into slow motion, to transfer one photographic style from one image to another, to change the weather in a photograph and to create photo-realistic faces from scratch. And there’s more where that come from.