Google is working on a new RawNeRF engine to outperform deep raw noise cancellation in low-light settings. This is how it will change night photography.
Shutterbugs will no longer have to worry about low light while taking their best photos. Google is working on a new AI image denoising technology that will effectively help photographers see in the dark. Basically, it can change the way you take low-light photos. This new Google Research has revealed an interesting technology using artificial intelligence to recreate a dark scene with powerful noise cancellation. Google research has called this new AI noise reduction tool RawNeRF in their open source project called MultiNeRF.
Google’s tool called RawNeRF is said to have unprecedented noise-cancelling power, being able to read images to add a greater level of detail in low-light settings. The team behind the project claims that it works better than any noise mitigation tools out there. This tool can change the game of the night photographyit added.
How Google’s NeRF tool will help in low light photography
The researchers explain in a Cornell University paper, “Neural light field (NeRF) is a technique for synthesizing a high-quality novel view from a collection of input images that have been shape. Like most view compositing methods, NeRF uses tone-mapped low dynamic range (LDR) as input; These images have been processed by a lossy camera pipeline to smooth out details, clip highlights, and distort the simple noise distribution of the raw sensor data. “
So this new tool not only acts as a denoiser, but can be used to control camera position, scene view, focus, exposure, and post-actual tone mapping. In a demonstration video by researcher Ben Mildenhall showing the use of NeRF in the dark. Mildenhall has a cell phone lit only by candlelight.
He used a smartphone shot of a candlelit table to demonstrate how the RawNeRF tool can help create a low-light shot. He used minimal post-processing and brightening while delivering the final image with more detail. “Although a single raw image appears significantly more noisy than a processed image onewe show that NeRF is very robust for the zero mean distribution of raw noise,” the researchers explain.
However, this tool is still in the research stage, but it will be interesting to see this Google’s AI technology used with our everyday gadgets in the future.
