Denoiser ML
Description
The Denoiser ML node employs sophisticated deep learning techniques to eliminate unwanted noise from images while preserving critical features. To train the denoising model, numerous images were utilized, resulting in a highly effective system. Additionally, the extracted noise can be seamlessly reintegrated into the image using the Grain Composite node.
Note: On Linux, the first time you apply the Denoiser, the ML models need to be compiled for your graphics card and resolution being used. The type of card and resolution will determine how long this process will take. Mac users should be running macOS 12.0 or higher for best performance when using the Denoiser ML node.
Node Group
Filter.
Controls
Models
Camera Noise
Reduces fine sensor noise from footage produced by contemporary cameras.
Compression+Noise
Reduces artifacts caused by compression and/or high ISO/ASA noise in low light scenes.
Amount
Sets the amount of denoising.
Outputs
Output
Outputs the denoised image.
Noise
Outputs only the removed noise. Use this output in conjunction with the Grain Composite node to composite the noise.