Class MaxineDenoiseVideoEffect
- Assembly
- VisioForge.Core.dll
AI-powered video denoising effect using NVIDIA Maxine technology. Implements the VisioForge.Core.Types.VideoEffects.VideoEffect.
public class MaxineDenoiseVideoEffect : VideoEffect, IVideoEffectInheritance
Implements
Inherited Members
Remarks
Available in: VideoCaptureCore, MediaPlayerCore, VideoEditCore (Windows only)
AI-powered effect using NVIDIA Maxine technology. Requires NVIDIA RTX GPU with Tensor cores.
The Maxine Denoise effect employs deep learning neural networks trained on extensive video datasets to intelligently remove noise while preserving important visual details and textures. Unlike traditional denoising filters that apply uniform smoothing or frequency-domain filtering, this AI-powered approach can distinguish between actual image details and various types of noise (Gaussian, compression artifacts, low-light noise) through learned pattern recognition.
The AI model analyzes spatial and temporal characteristics of the video to make intelligent decisions about which pixels represent noise versus legitimate image content. This results in cleaner video with better preserved edge definition, texture details, and natural appearance compared to conventional denoising methods. The effect is particularly effective for low-light footage, high-ISO camera captures, and compressed video streams where traditional methods often produce over-smoothed or blurred results.
System requirements:
- NVIDIA GeForce RTX 2060 or higher (Tensor cores required for AI inference)
- NVIDIA Maxine SDK installed with denoise models
- Minimum 4GB GPU memory recommended
- Windows 10/11 with latest NVIDIA drivers
Constructors
MaxineDenoiseVideoEffect(string, string, float)
Initializes a new instance of the VisioForge.Core.Types.VideoEffects.NvidiaMaxine.MaxineDenoiseVideoEffect class.
public MaxineDenoiseVideoEffect(string modelsDir, string name = "maxine_denoise", float strength = 0.7)Parameters
modelsDirstring-
The directory path containing the NVIDIA Maxine SDK denoise models.
namestring-
The unique name for this effect instance. Defaults to "maxine_denoise".
strengthfloat-
The denoising strength from 0.0 (weak) to 1.0 (strong). Defaults to 0.7 for balanced denoising.
Properties
ModelsDir
Gets the directory path where the NVIDIA Maxine SDK denoise models are located.
public string ModelsDir { get; }Property Value
Remarks
This directory must contain the denoise model files (.trt or .onnx) provided with the NVIDIA Maxine SDK. The models are optimized TensorRT engines that run on the GPU's Tensor cores for efficient AI inference.
Strength
Gets or sets the denoising strength, controlling the balance between noise removal and texture preservation.
public float Strength { get; set; }Property Value
Remarks
The AI model adapts its processing based on the strength value, with lower values preserving fine details like skin texture and fabric patterns, while higher values prioritize aggressive noise elimination. The optimal setting depends on the source video quality and the desired output characteristics.
Methods
GetEffectType()
Gets the type identifier for this video effect.
public VideoEffectType GetEffectType()Returns
- VideoEffectType
-
VisioForge.Core.Types.VideoEffects.VideoEffectType.MaxineDenoise indicating this is an NVIDIA Maxine AI denoising effect.