RealBasicVSROptions
Options & Configuration · AiDotNet.Video.Options
Configuration options for RealBasicVSR, real-world video super-resolution with an image-cleaning module.
Inheritance: ModelOptions → NeuralNetworkOptions → RealBasicVSROptions
For Beginners
RealBasicVSR first “cleans” each frame (removing noise, blur and compression artifacts), then upscales the cleaned frames with BasicVSR, so the artifacts are not spread from frame to frame.
How It Works
Defaults follow Chan et al., CVPR 2022, and the authors’ release (open-mmlab/mmagic, RealBasicVSRNet and configs/real_basicvsr/realbasicvsr_wogan-c64b20-2x30x8_8xb2-lr1e-4-300k_reds.py, the PSNR-oriented first stage): 64 channels, 20 propagation and 20 cleaning blocks, 15-frame clips, L1 output and cleaning losses against the unsharp-masked target, Adam (0.9, 0.99) at 1e-4 without a schedule, SPyNet frozen, and a 0.999 moving average of the weights for inference.
Constructors
| Constructor | Summary |
|---|---|
RealBasicVSROptions() |
Initializes a new instance with default values. |
RealBasicVSROptions(RealBasicVSROptions other) |
Initializes a new instance by copying from another instance. |
Properties
| Property | Summary |
|---|---|
OnnxModelOptions OnnxOptions { get; set; } |
Gets or sets the ONNX runtime options. |
bool SharpenTarget { get; set; } |
Gets or sets whether the losses score the unsharp-masked target (true, is_use_sharpened_gt_in_pixel). |
double CleaningLossWeight { get; set; } |
Gets or sets the weight of the L1 loss of the cleaned frames against the 4x area-downsampled target (1.0). |
double InferenceCleaningThreshold { get; set; } |
Gets or sets the mean absolute cleaning residue, in pixel levels 0-255, below which inference stops cleaning (1.5, the release’s test setting; training uses 255, i.e. |
double LearningRate { get; set; } |
Gets or sets Adam’s learning rate (1e-4). |
double PixelLossWeight { get; set; } |
Gets or sets the weight of the L1 loss on the output (1.0). |
double UnsharpThreshold { get; set; } |
Gets or sets the unsharp-masking residue threshold in pixel levels (10). |
double UnsharpWeight { get; set; } |
Gets or sets the unsharp-masking weight (0.5). |
double? EmaMomentum { get; set; } |
Gets or sets the momentum of the weights’ moving average used for inference (0.001); null disables it and inference uses the trained weights. |
int MidChannels { get; set; } |
Gets or sets the feature channels (64). |
int NumCleaningBlocks { get; set; } |
Gets or sets the residual blocks of the cleaning module (20). |
int NumFrames { get; set; } |
Gets or sets the frames of a training clip (15). |
int NumPropagationBlocks { get; set; } |
Gets or sets the residual blocks per propagation branch (20). |
int UnsharpKernelSize { get; set; } |
Gets or sets the unsharp-masking Gaussian size (51). |
string ModelPath { get; set; } |
Gets or sets the path to the ONNX model file. |