FastDVDNetOptions

Options & Configuration · AiDotNet.Video.Options

Configures FastDVDnet video-denoising training and architecture settings.

Inheritance: ModelOptions → NeuralNetworkOptions → ModelHyperparameterOptions → VideoHyperparameterOptions → FastDVDNetOptions

For Beginners

These settings control how FastDVDnet learns to remove noise by combining information from a five-frame temporal window.

How It Works

FastDVDnet uses two compact convolutional denoising stages without optical-flow estimation; the standard model uses 32 feature channels.

Reference: Tassano, Delon, and Veit, “FastDVDnet: Towards Real-Time Deep Video Denoising Without Flow Estimation,” 2020.

Constructors

Constructor Summary
FastDVDNetOptions() Initializes FastDVDnet options with the released training defaults.
FastDVDNetOptions(FastDVDNetOptions other) Initializes an independent copy of another FastDVDnet configuration.

Properties

Property Summary
double LearningRate { get; set; } Initial Adam learning rate.
int NumInputFrames { get; set; } Gets or sets the num input frames.

Methods

Method Summary
void Validate() Throws if a value this model requires has been left unset or is not positive.