VideoFlow<T>

Models & Types · AiDotNet.Video.Motion

VideoFlow multi-frame temporal cues for optical flow estimation.

Inheritance: NeuralNetworkBase<T> → VideoNeuralNetworkBase<T> → OpticalFlowBase<T> → VideoFlow<T>

For Beginners

VideoFlow extends optical flow estimation to video sequences by exploiting temporal continuity. It produces smoother, more consistent flow fields across consecutive frames.

How It Works

References:

  • Paper: “VideoFlow: Exploiting Temporal Cues for Multi-frame Optical Flow Estimation” (Shi et al., 2023)

VideoFlow exploits temporal cues from multiple frames simultaneously to improve optical flow estimation accuracy and temporal consistency.

Example

// Create a VideoFlow model for multi-frame temporal optical flow
var videoFlow = new VideoFlow<double>();
            
// Or configure with custom parameters
var architecture = new NeuralNetworkArchitecture<double>(
    inputType: InputType.ThreeDimensional,
    taskType: NeuralNetworkTaskType.Regression,
    inputHeight: 256, inputWidth: 256, inputDepth: 3, outputSize: 2);
var model = new VideoFlow<double>(architecture);

Constructors

Constructor Summary
VideoFlow() Initializes a new instance with default architecture settings.
VideoFlow(NeuralNetworkArchitecture<T> architecture, VideoFlowOptions options) Creates a new VideoFlow model for native training and inference.

Methods

Method Summary
ModelMetadata<T> GetModelMetadata()
Tensor<T> EstimateFlow(Tensor<T> frame0, Tensor<T> frame1)